Computer Organization and Design 2nd Edition

David A. pdf download

https://ebookname.com/product/computer-organization-and-
design-2nd-edition-david-a/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Computer Organization and Design The Hardware Software

Interface 3rd Edition David A. Patterson

https://ebookname.com/product/computer-organization-and-design-
the-hardware-software-interface-3rd-edition-david-a-patterson/

Digital Design and Computer Organization 1st Edition

Hassan A. Farhat

https://ebookname.com/product/digital-design-and-computer-
organization-1st-edition-hassan-a-farhat/

Structured computer organization 6ed. Edition Tanenbaum

A.S.

https://ebookname.com/product/structured-computer-
organization-6ed-edition-tanenbaum-a-s/

Molecular Epidemiology of Microorganisms Methods and

Protocols 1st Edition Michel Tibayrenc (Auth.)

https://ebookname.com/product/molecular-epidemiology-of-
microorganisms-methods-and-protocols-1st-edition-michel-
tibayrenc-auth/
I Could Tell You But Then You Would Have to be
Destroyed by Me Trevor Paglen

https://ebookname.com/product/i-could-tell-you-but-then-you-
would-have-to-be-destroyed-by-me-trevor-paglen/

Quadrature Domains and Their Applications The Harold S

Shapiro Anniversary Volume Operator Theory Advances and
Applications 15 1st Edition Peter Ebenfelt

https://ebookname.com/product/quadrature-domains-and-their-
applications-the-harold-s-shapiro-anniversary-volume-operator-
theory-advances-and-applications-15-1st-edition-peter-ebenfelt/

Inherit the holy mountain religion and the rise of

American environmentalism 1st Edition Stoll

https://ebookname.com/product/inherit-the-holy-mountain-religion-
and-the-rise-of-american-environmentalism-1st-edition-stoll/

Group Work With Persons With Disabilities 1st Edition

Bauman

https://ebookname.com/product/group-work-with-persons-with-
disabilities-1st-edition-bauman/

Ghettostadt Lodz and the Making of a Nazi City 1st

Edition Gordon J. Horwitz

https://ebookname.com/product/ghettostadt-lodz-and-the-making-of-
a-nazi-city-1st-edition-gordon-j-horwitz/
Mexican Labor Migrants and U S Immigration Policies
From Sojourner to Emigrant 1st Edition Florian K.
Kaufmann

https://ebookname.com/product/mexican-labor-migrants-and-u-s-
immigration-policies-from-sojourner-to-emigrant-1st-edition-
florian-k-kaufmann/
1 Fundamentals of
Computer Design 1

And now for something completely different.

Monty Python’s Flying Circus

 1.1 Introduction 1

1.2 The Task of a Computer Designer 3

1.3 Technology and Computer Usage Trends 6

1.4 Cost and Trends in Cost 8

1.5 Measuring and Reporting Performance 18

1.6 Quantitative Principles of Computer Design 29

1.7 Putting It All Together: The Concept of Memory Hierarchy 39

1.8 Fallacies and Pitfalls 44

1.9 Concluding Remarks 51

1.10 Historical Perspective and References 53

Exercises 60

1.1 Introduction
Computer technology has made incredible progress in the past half century. In
1945, there were no stored-program computers. Today, a few thousand dollars
will purchase a personal computer that has more performance, more main memo-
ry, and more disk storage than a computer bought in 1965 for $1 million. This
rapid rate of improvement has come both from advances in the technology used
to build computers and from innovation in computer design. While technological
improvements have been fairly steady, progress arising from better computer
architectures has been much less consistent. During the first 25 years of elec-
tronic computers, both forces made a major contribution; but beginning in about
1970, computer designers became largely dependent upon integrated circuit tech-
nology. During the 1970s, performance continued to improve at about 25% to
30% per year for the mainframes and minicomputers that dominated the industry.
The late 1970s saw the emergence of the microprocessor. The ability of the
microprocessor to ride the improvements in integrated circuit technology more
closely than the less integrated mainframes and minicomputers led to a higher
rate of improvement—roughly 35% growth per year in performance.
2 Chapter 1 Fundamentals of Computer Design

This growth rate, combined with the cost advantages of a mass-produced

microprocessor, led to an increasing fraction of the computer business being
based on microprocessors. In addition, two significant changes in the computer
marketplace made it easier than ever before to be commercially successful with a
new architecture. First, the virtual elimination of assembly language program-
ming reduced the need for object-code compatibility. Second, the creation of
standardized, vendor-independent operating systems, such as UNIX, lowered the
cost and risk of bringing out a new architecture. These changes made it possible
to successively develop a new set of architectures, called RISC architectures, in
the early 1980s. Since the RISC-based microprocessors reached the market in the
mid 1980s, these machines have grown in performance at an annual rate of over
50%. Figure 1.1 shows this difference in performance growth rates.

350
DEC Alpha

300

250 1.58x per year

200
SPECint rating

DEC Alpha
150
IBM Power2
DEC Alpha
100
1.35x per year
HP
9000
50 MIPS IBM
MIPS Power1
R3000
SUN4 R2000

0
4

5
8

9
19

19

19

19

19

19

19

19

19

19

19

19

Year

FIGURE 1.1 Growth in microprocessor performance since the mid 1980s has been substantially higher than in ear-
lier years. This chart plots the performance as measured by the SPECint benchmarks. Prior to the mid 1980s, micropro-
cessor performance growth was largely technology driven and averaged about 35% per year. The increase in growth since
then is attributable to more advanced architectural ideas. By 1995 this growth leads to more than a factor of five difference
in performance. Performance for floating-point-oriented calculations has increased even faster.
 1.2 The Task of a Computer Designer 3

The effect of this dramatic growth rate has been twofold. First, it has signifi-
cantly enhanced the capability available to computer users. As a simple example,
consider the highest-performance workstation announced in 1993, an IBM
Power-2 machine. Compared with a CRAY Y-MP supercomputer introduced in
1988 (probably the fastest machine in the world at that point), the workstation of-
fers comparable performance on many floating-point programs (the performance
for the SPEC floating-point benchmarks is similar) and better performance on in-
teger programs for a price that is less than one-tenth of the supercomputer!
Second, this dramatic rate of improvement has led to the dominance of micro-
processor-based computers across the entire range of the computer design. Work-
stations and PCs have emerged as major products in the computer industry.
Minicomputers, which were traditionally made from off-the-shelf logic or from
gate arrays, have been replaced by servers made using microprocessors. Main-
frames are slowly being replaced with multiprocessors consisting of small num-
bers of off-the-shelf microprocessors. Even high-end supercomputers are being
built with collections of microprocessors.
Freedom from compatibility with old designs and the use of microprocessor
technology led to a renaissance in computer design, which emphasized both ar-
chitectural innovation and efficient use of technology improvements. This renais-
sance is responsible for the higher performance growth shown in Figure 1.1—a
rate that is unprecedented in the computer industry. This rate of growth has com-
pounded so that by 1995, the difference between the highest-performance micro-
processors and what would have been obtained by relying solely on technology is
more than a factor of five. This text is about the architectural ideas and accom-
panying compiler improvements that have made this incredible growth rate possi-
ble. At the center of this dramatic revolution has been the development of a
quantitative approach to computer design and analysis that uses empirical obser-
vations of programs, experimentation, and simulation as its tools. It is this style
and approach to computer design that is reflected in this text.
Sustaining the recent improvements in cost and performance will require con-
tinuing innovations in computer design, and the authors believe such innovations
will be founded on this quantitative approach to computer design. Hence, this
book has been written not only to document this design style, but also to stimu-
late you to contribute to this progress.

1.2 The Task of a Computer Designer

The task the computer designer faces is a complex one: Determine what
attributes are important for a new machine, then design a machine to maximize
performance while staying within cost constraints. This task has many aspects,
including instruction set design, functional organization, logic design, and imple-
mentation. The implementation may encompass integrated circuit design,
4 Chapter 1 Fundamentals of Computer Design

packaging, power, and cooling. Optimizing the design requires familiarity with a
very wide range of technologies, from compilers and operating systems to logic
design and packaging.
In the past, the term computer architecture often referred only to instruction
set design. Other aspects of computer design were called implementation, often
insinuating that implementation is uninteresting or less challenging. The authors
believe this view is not only incorrect, but is even responsible for mistakes in the
design of new instruction sets. The architect’s or designer’s job is much more
than instruction set design, and the technical hurdles in the other aspects of the
project are certainly as challenging as those encountered in doing instruction set
design. This is particularly true at the present when the differences among in-
struction sets are small (see Appendix C).
In this book the term instruction set architecture refers to the actual programmer-
visible instruction set. The instruction set architecture serves as the boundary be-
tween the software and hardware, and that topic is the focus of Chapter 2. The im-
plementation of a machine has two components: organization and hardware. The
term organization includes the high-level aspects of a computer’s design, such as
the memory system, the bus structure, and the internal CPU (central processing
unit—where arithmetic, logic, branching, and data transfer are implemented)
design. For example, two machines with the same instruction set architecture but
different organizations are the SPARCstation-2 and SPARCstation-20. Hardware
is used to refer to the specifics of a machine. This would include the detailed
logic design and the packaging technology of the machine. Often a line of ma-
chines contains machines with identical instruction set architectures and nearly
identical organizations, but they differ in the detailed hardware implementation.
For example, two versions of the Silicon Graphics Indy differ in clock rate and in
detailed cache structure. In this book the word architecture is intended to cover
all three aspects of computer design—instruction set architecture, organization,
and hardware.
Computer architects must design a computer to meet functional requirements
as well as price and performance goals. Often, they also have to determine what
the functional requirements are, and this can be a major task. The requirements
may be specific features, inspired by the market. Application software often
drives the choice of certain functional requirements by determining how the ma-
chine will be used. If a large body of software exists for a certain instruction set
architecture, the architect may decide that a new machine should implement an
existing instruction set. The presence of a large market for a particular class of
applications might encourage the designers to incorporate requirements that
would make the machine competitive in that market. Figure 1.2 summarizes
some requirements that need to be considered in designing a new machine. Many
of these requirements and features will be examined in depth in later chapters.
Once a set of functional requirements has been established, the architect must
try to optimize the design. Which design choices are optimal depends, of course,
on the choice of metrics. The most common metrics involve cost and perfor-
 1.2 The Task of a Computer Designer 5

Functional requirements Typical features required or supported

Application area Target of computer
General purpose Balanced performance for a range of tasks (Ch 2,3,4,5)
Scientific High-performance floating point (App A,B)
Commercial Support for COBOL (decimal arithmetic); support for databases and transaction
processing (Ch 2,7)
Level of software compatibility Determines amount of existing software for machine
At programming language Most flexible for designer; need new compiler (Ch 2,8)
Object code or binary compatible Instruction set architecture is completely defined—little flexibility—but no in-
vestment needed in software or porting programs
Operating system requirements Necessary features to support chosen OS (Ch 5,7)
Size of address space Very important feature (Ch 5); may limit applications
Memory management Required for modern OS; may be paged or segmented (Ch 5)
Protection Different OS and application needs: page vs. segment protection (Ch 5)
Standards Certain standards may be required by marketplace
Floating point Format and arithmetic: IEEE, DEC, IBM (App A)
I/O bus For I/O devices: VME, SCSI, Fiberchannel (Ch 7)
Operating systems UNIX, DOS, or vendor proprietary
Networks Support required for different networks: Ethernet, ATM (Ch 6)
Programming languages Languages (ANSI C, Fortran 77, ANSI COBOL) affect instruction set (Ch 2)
FIGURE 1.2 Summary of some of the most important functional requirements an architect faces. The left-hand col-
umn describes the class of requirement, while the right-hand column gives examples of specific features that might be
needed. The right-hand column also contains references to chapters and appendices that deal with the specific issues.

mance. Given some application domain, the architect can try to quantify the per-
formance of the machine by a set of programs that are chosen to represent that
application domain. Other measurable requirements may be important in some
markets; reliability and fault tolerance are often crucial in transaction processing
environments. Throughout this text we will focus on optimizing machine cost/
performance.
In choosing between two designs, one factor that an architect must consider is
design complexity. Complex designs take longer to complete, prolonging time to
market. This means a design that takes longer will need to have higher perfor-
mance to be competitive. The architect must be constantly aware of the impact of
his design choices on the design time for both hardware and software.
In addition to performance, cost is the other key parameter in optimizing cost/
performance. In addition to cost, designers must be aware of important trends in
both the implementation technology and the use of computers. Such trends not
only impact future cost, but also determine the longevity of an architecture. The
next two sections discuss technology and cost trends.
6 Chapter 1 Fundamentals of Computer Design

1.3 Technology and Computer Usage Trends

If an instruction set architecture is to be successful, it must be designed to survive
changes in hardware technology, software technology, and application character-
istics. The designer must be especially aware of trends in computer usage and in
computer technology. After all, a successful new instruction set architecture may
last decades—the core of the IBM mainframe has been in use since 1964. An ar-
chitect must plan for technology changes that can increase the lifetime of a suc-
cessful machine.

Trends in Computer Usage

The design of a computer is fundamentally affected both by how it will be used
and by the characteristics of the underlying implementation technology. Changes
in usage or in implementation technology affect the computer design in different
ways, from motivating changes in the instruction set to shifting the payoff from
important techniques such as pipelining or caching.
Trends in software technology and how programs will use the machine have a
long-term impact on the instruction set architecture. One of the most important
software trends is the increasing amount of memory used by programs and their
data. The amount of memory needed by the average program has grown by a fac-
tor of 1.5 to 2 per year! This translates to a consumption of address bits at a rate
of approximately 1/2 bit to 1 bit per year. This rapid rate of growth is driven both
by the needs of programs as well as by the improvements in DRAM technology
that continually improve the cost per bit. Underestimating address-space growth
is often the major reason why an instruction set architecture must be abandoned.
(For further discussion, see Chapter 5 on memory hierarchy.)
Another important software trend in the past 20 years has been the replace-
ment of assembly language by high-level languages. This trend has resulted in a
larger role for compilers, forcing compiler writers and architects to work together
closely to build a competitive machine. Compilers have become the primary
interface between user and machine.
In addition to this interface role, compiler technology has steadily improved,
taking on newer functions and increasing the efficiency with which a program
can be run on a machine. This improvement in compiler technology has included
traditional optimizations, which we discuss in Chapter 2, as well as transforma-
tions aimed at improving pipeline behavior (Chapters 3 and 4) and memory sys-
tem behavior (Chapter 5). How to balance the responsibility for efficient
execution in modern processors between the compiler and the hardware contin-
ues to be one of the hottest architecture debates of the 1990s. Improvements in
compiler technology played a major role in making vector machines (Appendix
B) successful. The development of compiler technology for parallel machines is
likely to have a large impact in the future.
1.3 Technology and Computer Usage Trends 7

Trends in Implementation Technology

To plan for the evolution of a machine, the designer must be especially aware of
rapidly occurring changes in implementation technology. Three implementation
technologies, which change at a dramatic pace, are critical to modern implemen-
tations:
■ Integrated circuit logic technology—Transistor density increases by about
50% per year, quadrupling in just over three years. Increases in die size are less
predictable, ranging from 10% to 25% per year. The combined effect is a
growth rate in transistor count on a chip of between 60% and 80% per year. De-
vice speed increases nearly as fast; however, metal technology used for wiring
does not improve, causing cycle times to improve at a slower rate. We discuss
this further in the next section.
■ Semiconductor DRAM—Density increases by just under 60% per year, quadru-
pling in three years. Cycle time has improved very slowly, decreasing by about
one-third in 10 years. Bandwidth per chip increases as the latency decreases. In
addition, changes to the DRAM interface have also improved the bandwidth;
these are discussed in Chapter 5. In the past, DRAM (dynamic random-access
memory) technology has improved faster than logic technology. This differ-
ence has occurred because of reductions in the number of transistors per
DRAM cell and the creation of specialized technology for DRAMs. As the im-
provement from these sources diminishes, the density growth in logic technol-
ogy and memory technology should become comparable.
■ Magnetic disk technology—Recently, disk density has been improving by
about 50% per year, almost quadrupling in three years. Prior to 1990, density
increased by about 25% per year, doubling in three years. It appears that disk
technology will continue the faster density growth rate for some time to come.
Access time has improved by one-third in 10 years. This technology is central
to Chapter 6.
These rapidly changing technologies impact the design of a microprocessor
that may, with speed and technology enhancements, have a lifetime of five or
more years. Even within the span of a single product cycle (two years of design
and two years of production), key technologies, such as DRAM, change suffi-
ciently that the designer must plan for these changes. Indeed, designers often de-
sign for the next technology, knowing that when a product begins shipping in
volume that next technology may be the most cost-effective or may have perfor-
mance advantages. Traditionally, cost has decreased very closely to the rate at
which density increases.
These technology changes are not continuous but often occur in discrete steps.
For example, DRAM sizes are always increased by factors of four because of the
basic design structure. Thus, rather than doubling every 18 months, DRAM tech-
nology quadruples every three years. This stepwise change in technology leads to
8 Chapter 1 Fundamentals of Computer Design

thresholds that can enable an implementation technique that was previously im-
possible. For example, when MOS technology reached the point where it could
put between 25,000 and 50,000 transistors on a single chip in the early 1980s, it
became possible to build a 32-bit microprocessor on a single chip. By eliminating
chip crossings within the processor, a dramatic increase in cost/performance was
possible. This design was simply infeasible until the technology reached a certain
point. Such technology thresholds are not rare and have a significant impact on a
wide variety of design decisions.

1.4 Cost and Trends in Cost

Although there are computer designs where costs tend to be ignored—
specifically supercomputers—cost-sensitive designs are of growing importance.
Indeed, in the past 15 years, the use of technology improvements to achieve low-
er cost, as well as increased performance, has been a major theme in the comput-
er industry. Textbooks often ignore the cost half of cost/performance because
costs change, thereby dating books, and because the issues are complex. Yet an
understanding of cost and its factors is essential for designers to be able to make
intelligent decisions about whether or not a new feature should be included in de-
signs where cost is an issue. (Imagine architects designing skyscrapers without
any information on costs of steel beams and concrete.) This section focuses on
cost, specifically on the components of cost and the major trends. The Exercises
and Examples use specific cost data that will change over time, though the basic
determinants of cost are less time sensitive.
Entire books are written about costing, pricing strategies, and the impact of
volume. This section can only introduce you to these topics by discussing some
of the major factors that influence cost of a computer design and how these fac-
tors are changing over time.

The Impact of Time, Volume, Commodization,

and Packaging
The cost of a manufactured computer component decreases over time even with-
out major improvements in the basic implementation technology. The underlying
principle that drives costs down is the learning curve—manufacturing costs de-
crease over time. The learning curve itself is best measured by change in yield—
the percentage of manufactured devices that survives the testing procedure.
Whether it is a chip, a board, or a system, designs that have twice the yield will
have basically half the cost. Understanding how the learning curve will improve
yield is key to projecting costs over the life of the product. As an example of the
learning curve in action, the cost per megabyte of DRAM drops over the long
term by 40% per year. A more dramatic version of the same information is shown
 1.4 Cost and Trends in Cost 9

in Figure 1.3, where the cost of a new DRAM chip is depicted over its lifetime.
Between the start of a project and the shipping of a product, say two years, the
cost of a new DRAM drops by a factor of between five and 10 in constant dollars.
Since not all component costs change at the same rate, designs based on project-
ed costs result in different cost/performance trade-offs than those using current
costs. The caption of Figure 1.3 discusses some of the long-term trends in DRAM
cost.

80

16 MB
70

60

50

4 MB
1 MB
Dollars per 40
DRAM chip 256 KB

30
Final chip cost
64 KB
20

10 16 KB

0
8

5
7

9
19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19
Year

FIGURE 1.3 Prices of four generations of DRAMs over time in 1977 dollars, showing the learning curve at work. A
1977 dollar is worth about $2.44 in 1995; most of this inflation occurred in the period of 1977–82, during which the value
changed to $1.61. The cost of a megabyte of memory has dropped incredibly during this period, from over $5000 in 1977 to
just over $6 in 1995 (in 1977 dollars)! Each generation drops in constant dollar price by a factor of 8 to 10 over its lifetime.
The increasing cost of fabrication equipment for each new generation has led to slow but steady increases in both the start-
ing price of a technology and the eventual, lowest price. Periods when demand exceeded supply, such as 1987–88 and
1992–93, have led to temporary higher pricing, which shows up as a slowing in the rate of price decrease.
10 Chapter 1 Fundamentals of Computer Design

Volume is a second key factor in determining cost. Increasing volumes affect

cost in several ways. First, they decrease the time needed to get down the learning
curve, which is partly proportional to the number of systems (or chips) manufac-
tured. Second, volume decreases cost, since it increases purchasing and manufac-
turing efficiency. As a rule of thumb, some designers have estimated that cost
decreases about 10% for each doubling of volume. Also, volume decreases the
amount of development cost that must be amortized by each machine, thus
allowing cost and selling price to be closer. We will return to the other factors in-
fluencing selling price shortly.
Commodities are products that are sold by multiple vendors in large volumes
and are essentially identical. Virtually all the products sold on the shelves of gro-
cery stores are commodities, as are standard DRAMs, small disks, monitors, and
keyboards. In the past 10 years, much of the low end of the computer business
has become a commodity business focused on building IBM-compatible PCs.
There are a variety of vendors that ship virtually identical products and are highly
competitive. Of course, this competition decreases the gap between cost and sell-
ing price, but it also decreases cost. This occurs because a commodity market has
both volume and a clear product definition. This allows multiple suppliers to
compete in building components for the commodity product. As a result, the
overall product cost is lower because of the competition among the suppliers of
the components and the volume efficiencies the suppliers can achieve.

Cost of an Integrated Circuit

Why would a computer architecture book have a section on integrated circuit
costs? In an increasingly competitive computer marketplace where standard
parts—disks, DRAMs, and so on—are becoming a significant portion of any sys-
tem’s cost, integrated circuit costs are becoming a greater portion of the cost that
varies between machines, especially in the high-volume, cost-sensitive portion of
the market. Thus computer designers must understand the costs of chips to under-
stand the costs of current computers. We follow here the U.S. accounting ap-
proach to the costs of chips.
While the costs of integrated circuits have dropped exponentially, the basic
procedure of silicon manufacture is unchanged: A wafer is still tested and
chopped into dies that are packaged (see Figures 1.4 and 1.5). Thus the cost of a
packaged integrated circuit is
Cost of die + Cost of testing die + Cost of packaging and final test
Cost of integrated circuit =
Final test yield

In this section, we focus on the cost of dies, summarizing the key issues in testing
and packaging at the end. A longer discussion of the testing costs and packaging
costs appears in the Exercises.
 1.4 Cost and Trends in Cost 11

FIGURE 1.4 Photograph of an 8-inch wafer containing Intel Pentium microprocessors. The die size is 480.7 mm2
and the total number of dies is 63. (Courtesy Intel.)

FIGURE 1.5 Photograph of an 8-inch wafer containing PowerPC 601 microprocessors. The die size is 122 mm2. The
number of dies on the wafer is 200 after subtracting the test dies (the odd-looking dies that are scattered around). (Courtesy
IBM.)
12 Chapter 1 Fundamentals of Computer Design

To learn how to predict the number of good chips per wafer requires first
learning how many dies fit on a wafer and then learning how to predict the per-
centage of those that will work. From there it is simple to predict cost:
Cost of wafer
Cost of die = ---------------------------------------------------------------
Dies per wafer × Die yield
The most interesting feature of this first term of the chip cost equation is its sensi-
tivity to die size, shown below.
The number of dies per wafer is basically the area of the wafer divided by the
area of the die. It can be more accurately estimated by
2
π × ( Wafer diameter/2 ) π × Wafer diameter
Dies per wafer = ----------------------------------------------------------- – -----------------------------------------------
Die area 2 × Die area

The first term is the ratio of wafer area (πr2) to die area. The second compensates
for the “square peg in a round hole” problem—rectangular dies near the periphery
of round wafers. Dividing the circumference (πd) by the diagonal of a square die is
approximately the number of dies along the edge. For example, a wafer 20 cm (≈ 8
inch) in diameter produces 3.14 × 100 – ( 3.14 × 20 ⁄ 1.41 ) = 269 1-cm dies.

EXAMPLE Find the number of dies per 20-cm wafer for a die that is 1.5 cm on a side.

ANSWER The total die area is 2.25 cm2. Thus

2
π × ( 20 ⁄ 2 ) π × 20 314 62.8
Dies per wafer = ------------------------------ – ------------------------ = ---------- – ---------- = 110
2.25 2 × 2.25 2.25 2.12
■

But this only gives the maximum number of dies per wafer. The critical ques-
tion is, What is the fraction or percentage of good dies on a wafer number, or the
die yield? A simple empirical model of integrated circuit yield, which assumes
that defects are randomly distributed over the wafer and that yield is inversely
proportional to the complexity of the fabrication process, leads to the following:

Defects per unit area × Die area –α

Die yield = Wafer yield ×  1 + ----------------------------------------------------------------------------
 α 

where wafer yield accounts for wafers that are completely bad and so need not be
tested. For simplicity, we’ll just assume the wafer yield is 100%. Defects per unit
area is a measure of the random and manufacturing defects that occur. In 1995,
these values typically range between 0.6 and 1.2 per square centimeter, depend-
ing on the maturity of the process (recall the learning curve, mentioned earlier).
Lastly, α is a parameter that corresponds roughly to the number of masking lev-
els, a measure of manufacturing complexity, critical to die yield. For today’s mul-
tilevel metal CMOS processes, a good estimate is α = 3.0.
 1.4 Cost and Trends in Cost 13

EXAMPLE Find the die yield for dies that are 1 cm on a side and 1.5 cm on a side,
assuming a defect density of 0.8 per cm2.

ANSWER The total die areas are 1 cm2 and 2.25 cm2. For the smaller die the yield is

0.8 × 1 – 3
Die yield =  1 + ---------------- = 0.49
 3 

For the larger die, it is

0.8 × 2.25 – 3
Die yield =  1 + ------------------------ = 0.24
 3 
■

The bottom line is the number of good dies per wafer, which comes from mul-
tiplying dies per wafer by die yield. The examples above predict 132 good 1-cm2
dies from the 20-cm wafer and 26 good 2.25-cm2 dies. Most high-end micro-
processors fall between these two sizes, with some being as large as 2.75 cm2 in
1995. Low-end processors are sometimes as small as 0.8 cm2, while processors
used for embedded control (in printers, automobiles, etc.) are often just 0.5 cm2.
(Figure 1.22 on page 63 in the Exercises shows the die size and technology for sev-
eral current microprocessors.) Occasionally dies become pad limited: the amount
of die area is determined by the perimeter rather than the logic in the interior. This
may lead to a higher yield, since defects in empty silicon are less serious!
Processing a 20-cm-diameter wafer in a leading-edge technology with 3–4
metal layers costs between $3000 and $4000 in 1995. Assuming a processed wa-
fer cost of $3500, the cost of the 1-cm2 die is around $27, while the cost per die
of the 2.25-cm2 die is about $140, or slightly over 5 times the cost for a die that is
2.25 times larger.
What should a computer designer remember about chip costs? The manufac-
turing process dictates the wafer cost, wafer yield, α, and defects per unit area, so
the sole control of the designer is die area. Since α is typically 3 for the advanced
processes in use today, die costs are proportional to the fourth (or higher) power
of the die area:
Cost of die = f (Die area4)

The computer designer affects die size, and hence cost, both by what functions
are included on or excluded from the die and by the number of I/O pins.
Before we have a part that is ready for use in a computer, the part must be
tested (to separate the good dies from the bad), packaged, and tested again after
packaging. These steps all add costs. These processes and their contribution to
cost are discussed and evaluated in Exercise 1.8.
14 Chapter 1 Fundamentals of Computer Design

Distribution of Cost in a System: An Example

To put the costs of silicon in perspective, Figure 1.6 shows the approximate cost
breakdown for a color desktop machine in the late 1990s. While costs for units
like DRAMs will surely drop over time from those in Figure 1.6, costs for units
whose prices have already been cut, like displays and cabinets, will change very
little. Furthermore, we can expect that future machines will have larger memories
and disks, meaning that prices drop more slowly than the technology improve-
ment.
The processor subsystem accounts for only 6% of the overall cost. Although in
a mid-range or high-end design this number would be larger, the overall break-
down across major subsystems is likely to be similar.

System Subsystem Fraction of total

Cabinet Sheet metal, plastic 1%
Power supply, fans 2%
Cables, nuts, bolts 1%
Shipping box, manuals 0%
Subtotal 4%
Processor board Processor 6%
DRAM (64 MB) 36%
Video system 14%
I/O system 3%
Printed circuit board 1%
Subtotal 60%
I/O devices Keyboard and mouse 1%
Monitor 22%
Hard disk (1 GB) 7%
DAT drive 6%
Subtotal 36%
FIGURE 1.6 Estimated distribution of costs of the components in a low-end, late
1990s color desktop workstation assuming 100,000 units. Notice that the largest single
item is memory! Costs for a high-end PC would be similar, except that the amount of memory
might be 16–32 MB rather than 64 MB. This chart is based on data from Andy Bechtolsheim
of Sun Microsystems, Inc. Touma [1993] discusses workstation costs and pricing.

Cost Versus Price—Why They Differ and By How Much

Costs of components may confine a designer’s desires, but they are still far from
representing what the customer must pay. But why should a computer architec-
ture book contain pricing information? Cost goes through a number of changes
1.4 Cost and Trends in Cost 15

before it becomes price, and the computer designer should understand how a de-
sign decision will affect the potential selling price. For example, changing cost
by $1000 may change price by $3000 to $4000. Without understanding the rela-
tionship of cost to price the computer designer may not understand the impact on
price of adding, deleting, or replacing components. The relationship between
price and volume can increase the impact of changes in cost, especially at the low
end of the market. Typically, fewer computers are sold as the price increases. Fur-
thermore, as volume decreases, costs rise, leading to further increases in price.
Thus, small changes in cost can have a larger than obvious impact. The relation-
ship between cost and price is a complex one with entire books written on the
subject. The purpose of this section is to give you a simple introduction to what
factors determine price and typical ranges for these factors.
The categories that make up price can be shown either as a tax on cost or as a
percentage of the price. We will look at the information both ways. These differ-
ences between price and cost also depend on where in the computer marketplace
a company is selling. To show these differences, Figures 1.7 and 1.8 on page 16
show how the difference between cost of materials and list price is decomposed,
with the price increasing from left to right as we add each type of overhead.
Direct costs refer to the costs directly related to making a product. These in-
clude labor costs, purchasing components, scrap (the leftover from yield), and
warranty, which covers the costs of systems that fail at the customer’s site during
the warranty period. Direct cost typically adds 20% to 40% to component cost.
Service or maintenance costs are not included because the customer typically
pays those costs, although a warranty allowance may be included here or in gross
margin, discussed next.
The next addition is called the gross margin, the company’s overhead that can-
not be billed directly to one product. This can be thought of as indirect cost. It in-
cludes the company’s research and development (R&D), marketing, sales,
manufacturing equipment maintenance, building rental, cost of financing, pretax
profits, and taxes. When the component costs are added to the direct cost and
gross margin, we reach the average selling price—ASP in the language of
MBAs—the money that comes directly to the company for each product sold.
The gross margin is typically 20% to 55% of the average selling price, depending
on the uniqueness of the product. Manufacturers of low-end PCs generally have
lower gross margins for several reasons. First, their R&D expenses are lower.
Second, their cost of sales is lower, since they use indirect distribution (by mail,
phone order, or retail store) rather than salespeople. Third, because their products
are less unique, competition is more intense, thus forcing lower prices and often
lower profits, which in turn lead to a lower gross margin.
List price and average selling price are not the same. One reason for this is that
companies offer volume discounts, lowering the average selling price. Also, if the
product is to be sold in retail stores, as personal computers are, stores want to
keep 40% to 50% of the list price for themselves. Thus, depending on the distri-
bution system, the average selling price is typically 50% to 75% of the list price.
16 Chapter 1 Fundamentals of Computer Design

List
price

Average
33.3% discount

Average
selling
price

50%
Gross Gross
margin 33.3% margin

25% Direct costs 12.5% Direct costs 8.3% Direct costs

Component Component Component Component

100% 37.5% costs 25%
costs 75% costs costs

Add 33% for Add 100% for Add 50% for

direct costs gross margin average discount

FIGURE 1.7 The components of price for a mid-range product in a workstation com-
pany. Each increase is shown along the bottom as a tax on the prior price. The percentages
of the new price for all elements are shown on the left of each column.

List
price

45%
Average
discount

Average
selling
price Gross Gross
25% margin 14%
margin

25% Direct costs 19% Direct costs 10% Direct costs

Component Component 56% Component Component

100% 31%
costs 75% costs costs costs

Add 33% for Add 33% for Add 80% for

direct costs gross margin average discount

FIGURE 1.8 The components of price for a desktop product in a personal computer
company. A larger average discount is used because of indirect selling, and a lower gross
margin is required.
1.4 Cost and Trends in Cost 17

As we said, pricing is sensitive to competition: A company may not be able to

sell its product at a price that includes the desired gross margin. In the worst case,
the price must be significantly reduced, lowering gross margin until profit be-
comes negative! A company striving for market share can reduce price and profit
to increase the attractiveness of its products. If the volume grows sufficiently,
costs can be reduced. Remember that these relationships are extremely complex
and to understand them in depth would require an entire book, as opposed to one
section in one chapter. For example, if a company cuts prices, but does not obtain
a sufficient growth in product volume, the chief impact will be lower profits.
Many engineers are surprised to find that most companies spend only 4% (in
the commodity PC business) to 12% (in the high-end server business) of their in-
come on R&D, which includes all engineering (except for manufacturing and
field engineering). This is a well-established percentage that is reported in com-
panies’ annual reports and tabulated in national magazines, so this percentage is
unlikely to change over time.
The information above suggests that a company uniformly applies fixed-
overhead percentages to turn cost into price, and this is true for many companies.
But another point of view is that R&D should be considered an investment. Thus
an investment of 4% to 12% of income means that every $1 spent on R&D should
lead to $8 to $25 in sales. This alternative point of view then suggests a different
gross margin for each product depending on the number sold and the size of the
investment.
Large, expensive machines generally cost more to develop—a machine cost-
ing 10 times as much to manufacture may cost many times as much to develop.
Since large, expensive machines generally do not sell as well as small ones, the
gross margin must be greater on the big machines for the company to maintain a
profitable return on its investment. This investment model places large machines
in double jeopardy—because there are fewer sold and they require larger R&D
costs—and gives one explanation for a higher ratio of price to cost versus smaller
machines.
The issue of cost and cost/performance is a complex one. There is no single
target for computer designers. At one extreme, high-performance design spares
no cost in achieving its goal. Supercomputers have traditionally fit into this cate-
gory. At the other extreme is low-cost design, where performance is sacrificed to
achieve lowest cost. Computers like the IBM PC clones belong here. Between
these extremes is cost/performance design, where the designer balances cost ver-
sus performance. Most of the workstation manufacturers operate in this region. In
the past 10 years, as computers have downsized, both low-cost design and cost/
performance design have become increasingly important. Even the supercom-
puter manufacturers have found that cost plays an increasing role. This section
has introduced some of the most important factors in determining cost; the next
section deals with performance.
18 Chapter 1 Fundamentals of Computer Design

1.5 Measuring and Reporting Performance

When we say one computer is faster than another, what do we mean? The com-
puter user may say a computer is faster when a program runs in less time, while
the computer center manager may say a computer is faster when it completes
more jobs in an hour. The computer user is interested in reducing response
time—the time between the start and the completion of an event—also referred to
as execution time. The manager of a large data processing center may be interest-
ed in increasing throughput—the total amount of work done in a given time.
In comparing design alternatives, we often want to relate the performance of
two different machines, say X and Y. The phrase “X is faster than Y” is used here
to mean that the response time or execution time is lower on X than on Y for the
given task. In particular, “X is n times faster than Y” will mean

Execution time Y
---------------------------------------- = n
Execution time X

Since execution time is the reciprocal of performance, the following relationship

holds:
1
----------------------------------
Execution time Y Performance Y Performance X
n = ---------------------------------------- = ----------------------------------- = ----------------------------------
Execution time X 1 Performance Y
----------------------------------
Performance X

The phrase “the throughput of X is 1.3 times higher than Y” signifies here that
the number of tasks completed per unit time on machine X is 1.3 times the num-
ber completed on Y.
Because performance and execution time are reciprocals, increasing perfor-
mance decreases execution time. To help avoid confusion between the terms
increasing and decreasing, we usually say “improve performance” or “improve
execution time” when we mean increase performance and decrease execution
time.
Whether we are interested in throughput or response time, the key measure-
ment is time: The computer that performs the same amount of work in the least
time is the fastest. The difference is whether we measure one task (response time)
or many tasks (throughput). Unfortunately, time is not always the metric quoted
in comparing the performance of computers. A number of popular measures have
been adopted in the quest for a easily understood, universal measure of computer
performance, with the result that a few innocent terms have been shanghaied
from their well-defined environment and forced into a service for which they
were never intended. The authors’ position is that the only consistent and reliable
measure of performance is the execution time of real programs, and that all pro-
posed alternatives to time as the metric or to real programs as the items measured
1.5 Measuring and Reporting Performance 19

have eventually led to misleading claims or even mistakes in computer design.

The dangers of a few popular alternatives are shown in Fallacies and Pitfalls,
section 1.8.

Measuring Performance
Even execution time can be defined in different ways depending on what we
count. The most straightforward definition of time is called wall-clock time, re-
sponse time, or elapsed time, which is the latency to complete a task, including
disk accesses, memory accesses, input/output activities, operating system over-
head—everything. With multiprogramming the CPU works on another program
while waiting for I/O and may not necessarily minimize the elapsed time of one
program. Hence we need a term to take this activity into account. CPU time rec-
ognizes this distinction and means the time the CPU is computing, not including
the time waiting for I/O or running other programs. (Clearly the response time
seen by the user is the elapsed time of the program, not the CPU time.) CPU time
can be further divided into the CPU time spent in the program, called user CPU
time, and the CPU time spent in the operating system performing tasks requested
by the program, called system CPU time.
These distinctions are reflected in the UNIX time command, which returns
four measurements when applied to an executing program:
90.7u 12.9s 2:39 65%

User CPU time is 90.7 seconds, system CPU time is 12.9 seconds, elapsed time is
2 minutes and 39 seconds (159 seconds), and the percentage of elapsed time that
is CPU time is (90.7 + 12.9)/159 or 65%. More than a third of the elapsed time in
this example was spent waiting for I/O or running other programs or both. Many
measurements ignore system CPU time because of the inaccuracy of operating
systems’ self-measurement (the above inaccurate measurement came from UNIX)
and the inequity of including system CPU time when comparing performance be-
tween machines with differing system codes. On the other hand, system code on
some machines is user code on others, and no program runs without some operat-
ing system running on the hardware, so a case can be made for using the sum of
user CPU time and system CPU time.
In the present discussion, a distinction is maintained between performance
based on elapsed time and that based on CPU time. The term system performance
is used to refer to elapsed time on an unloaded system, while CPU performance
refers to user CPU time on an unloaded system. We will concentrate on CPU per-
formance in this chapter.
20 Chapter 1 Fundamentals of Computer Design

Choosing Programs to Evaluate Performance

Dhrystone does not use floating point. Typical programs don’t …

Rick Richardson, Clarification of Dhrystone (1988)

This program is the result of extensive research to determine the instruction mix
of a typical Fortran program. The results of this program on different machines
should give a good indication of which machine performs better under a typical
load of Fortran programs. The statements are purposely arranged to defeat opti-
mizations by the compiler.
H. J. Curnow and B. A. Wichmann [1976], Comments in the Whetstone Benchmark

A computer user who runs the same programs day in and day out would be the
perfect candidate to evaluate a new computer. To evaluate a new system the user
would simply compare the execution time of her workload—the mixture of pro-
grams and operating system commands that users run on a machine. Few are in
this happy situation, however. Most must rely on other methods to evaluate ma-
chines and often other evaluators, hoping that these methods will predict per-
formance for their usage of the new machine. There are four levels of programs
used in such circumstances, listed below in decreasing order of accuracy of pre-
diction.

1. Real programs—While the buyer may not know what fraction of time is spent
on these programs, she knows that some users will run them to solve real prob-
lems. Examples are compilers for C, text-processing software like TeX, and CAD
tools like Spice. Real programs have input, output, and options that a user can se-
lect when running the program.
2. Kernels—Several attempts have been made to extract small, key pieces from
real programs and use them to evaluate performance. Livermore Loops and Lin-
pack are the best known examples. Unlike real programs, no user would run kernel
programs, for they exist solely to evaluate performance. Kernels are best used to
isolate performance of individual features of a machine to explain the reasons for
differences in performance of real programs.
3. Toy benchmarks—Toy benchmarks are typically between 10 and 100 lines of
code and produce a result the user already knows before running the toy program.
Programs like Sieve of Eratosthenes, Puzzle, and Quicksort are popular because
they are small, easy to type, and run on almost any computer. The best use of such
programs is beginning programming assignments.
4. Synthetic benchmarks—Similar in philosophy to kernels, synthetic bench-
marks try to match the average frequency of operations and operands of a large set
of programs. Whetstone and Dhrystone are the most popular synthetic benchmarks.
1.5 Measuring and Reporting Performance 21

A description of these benchmarks and some of their flaws appears in section 1.8
on page 44. No user runs synthetic benchmarks, because they don’t compute any-
thing a user could want. Synthetic benchmarks are, in fact, even further removed
from reality because kernel code is extracted from real programs, while synthetic
code is created artificially to match an average execution profile. Synthetic bench-
marks are not even pieces of real programs, while kernels might be.

Because computer companies thrive or go bust depending on price/perfor-

mance of their products relative to others in the marketplace, tremendous re-
sources are available to improve performance of programs widely used in
evaluating machines. Such pressures can skew hardware and software engineer-
ing efforts to add optimizations that improve performance of synthetic programs,
toy programs, kernels, and even real programs. The advantage of the last of these
is that adding such optimizations is more difficult in real programs, though not
impossible. This fact has caused some benchmark providers to specify the rules
under which compilers must operate, as we will see shortly.

Benchmark Suites
Recently, it has become popular to put together collections of benchmarks to try
to measure the performance of processors with a variety of applications. Of
course, such suites are only as good as the constituent individual benchmarks.
Nonetheless, a key advantage of such suites is that the weakness of any one
benchmark is lessened by the presence of the other benchmarks. This is especial-
ly true if the methods used for summarizing the performance of the benchmark
suite reflect the time to run the entire suite, as opposed to rewarding performance
increases on programs that may be defeated by targeted optimizations. In the re-
mainder of this section, we discuss the strengths and weaknesses of different
methods for summarizing performance.
Benchmark suites are made of collections of programs, some of which may be
kernels, but many of which are typically real programs. Figure 1.9 describes the
programs in the popular SPEC92 benchmark suite used to characterize perfor-
mance in the workstation and server markets.The programs in SPEC92 vary from
collections of kernels (nasa7) to small, program fragments (tomcatv, ora, alvinn,
swm256) to applications of varying size (spice2g6, gcc, compress). We will see
data on many of these programs throughout this text. In the next subsection, we
show how a SPEC92 report describes the machine, compiler, and OS configura-
tion, while in section 1.8 we describe some of the pitfalls that have occurred in
attempting to develop the benchmark suite and to prevent the benchmark circum-
vention that makes the results not useful for comparing performance among
machines.
22 Chapter 1 Fundamentals of Computer Design

Benchmark Source Lines of code Description

espresso C 13,500 Minimizes Boolean functions.
li C 7,413 A lisp interpreter written in C that solves the 8-queens problem.
eqntott C 3,376 Translates a Boolean equation into a truth table.
compress C 1,503 Performs data compression on a 1-MB file using Lempel-Ziv
coding.
sc C 8,116 Performs computations within a UNIX spreadsheet.
gcc C 83,589 Consists of the GNU C compiler converting preprocessed files into
optimized Sun-3 machine code.
spice2g6 FORTRAN 18,476 Circuit simulation package that simulates a small circuit.
doduc FORTRAN 5,334 A Monte Carlo simulation of a nuclear reactor component.
mdljdp2 FORTRAN 4,458 A chemical application that solves equations of motion for a model
of 500 atoms. This is similar to modeling a structure of liquid argon.
wave5 FORTRAN 7,628 A two-dimensional electromagnetic particle-in-cell simulation used
to study various plasma phenomena. Solves equations of motion on
a mesh involving 500,000 particles on 50,000 grid points for 5 time
steps.
tomcatv FORTRAN 195 A mesh generation program, which is highly vectorizable.
ora FORTRAN 535 Traces rays through optical systems of spherical and plane surfaces.
mdljsp2 FORTRAN 3,885 Same as mdljdp2, but single precision.
alvinn C 272 Simulates training of a neural network. Uses single precision.
ear C 4,483 An inner ear model that filters and detects various sounds and
generates speech signals. Uses single precision.
swm256 FORTRAN 487 A shallow water model that solves shallow water equations using
finite difference equations with a 256 × 256 grid. Uses single
precision.
su2cor FORTRAN 2,514 Computes masses of elementary particles from Quark-Gluon theory.
hydro2d FORTRAN 4,461 An astrophysics application program that solves hydrodynamical
Navier Stokes equations to compute galactical jets.
nasa7 FORTRAN 1,204 Seven kernels do matrix manipulation, FFTs, Gaussian elimination,
vortices creation.
fpppp FORTRAN 2,718 A quantum chemistry application program used to calculate two
electron integral derivatives.
FIGURE 1.9 The programs in the SPEC92 benchmark suites. The top six entries are the integer-oriented programs,
from which the SPECint92 performance is computed. The bottom 14 are the floating-point-oriented benchmarks from which
the SPECfp92 performance is computed.The floating-point programs use double precision unless stated otherwise. The
amount of nonuser CPU activity varies from none (for most of the FP benchmarks) to significant (for programs like gcc and
compress). In the performance measurements in this text, we use the five integer benchmarks (excluding sc) and five FP
benchmarks: doduc, mdljdp2, ear, hydro2d, and su2cor.
1.5 Measuring and Reporting Performance 23

Reporting Performance Results

The guiding principle of reporting performance measurements should be repro-
ducibility—list everything another experimenter would need to duplicate the re-
sults. Compare descriptions of computer performance found in refereed scientific
journals to descriptions of car performance found in magazines sold at supermar-
kets. Car magazines, in addition to supplying 20 performance metrics, list all op-
tional equipment on the test car, the types of tires used in the performance test,
and the date the test was made. Computer journals may have only seconds of exe-
cution labeled by the name of the program and the name and model of the com-
puter—spice takes 187 seconds on an IBM RS/6000 Powerstation 590. Left to
the reader’s imagination are program input, version of the program, version of
compiler, optimizing level of compiled code, version of operating system,
amount of main memory, number and types of disks, version of the CPU—all of
which make a difference in performance. In other words, car magazines have
enough information about performance measurements to allow readers to dupli-
cate results or to question the options selected for measurements, but computer
journals often do not!
A SPEC benchmark report requires a fairly complete description of the ma-
chine, the compiler flags, as well as the publication of both the baseline and opti-
mized results. As an example, Figure 1.10 shows portions of the SPECfp92
report for an IBM RS/6000 Powerstation 590. In addition to hardware, software,
and baseline tuning parameter descriptions, a SPEC report contains the actual
performance times, shown both in tabular form and as a graph.
The importance of performance on the SPEC benchmarks motivated vendors
to add many benchmark-specific flags when compiling SPEC programs; these
flags often caused transformations that would be illegal on many programs or
would slow down performance on others. To restrict this process and increase the
significance of the SPEC results, the SPEC organization created a baseline per-
formance measurement in addition to the optimized performance measurement.
Baseline performance restricts the vendor to one compiler and one set of flags for
all the programs in the same language (C or FORTRAN). Figure 1.10 shows the
parameters for the baseline performance; in section 1.8, Fallacies and Pitfalls,
we’ll see the tuning parameters for the optimized performance runs on this
machine.

Comparing and Summarizing Performance

Comparing performance of computers is rarely a dull event, especially when the
designers are involved. Charges and countercharges fly across the Internet; one is
accused of underhanded tactics and the other of misleading statements. Since ca-
reers sometimes depend on the results of such performance comparisons, it is un-
derstandable that the truth is occasionally stretched. But more frequently
discrepancies can be explained by differing assumptions or lack of information.
24 Chapter 1 Fundamentals of Computer Design

Hardware Software
Model number Powerstation 590 O/S and version AIX version 3.2.5
CPU 66.67 MHz POWER2 Compilers and version C SET++ for AIX C/C++ version 2.1
XL FORTRAN/6000 version 3.1
FPU Integrated Other software See below
Number of CPUs 1 File system type AIX/JFS
Primary cache 32KBI+256KBD off chip System state Single user
Secondary cache None
Other cache None
Memory 128 MB
Disk subsystem 2x2.0 GB
Other hardware None
SPECbase_fp92 tuning parameters/notes/summary of changes:
FORTRAN flags: -O3 -qarch=pwrx -qhsflt -qnofold -bnso -BI:/lib/syscalss.exp
C flags: -O3 -qarch=pwrx -Q -qtune=pwrx -qhssngl -bnso -bI:/lib/syscalls.exp
FIGURE 1.10 The machine, software, and baseline tuning parameters for the SPECfp92 report on an IBM RS/6000
Powerstation 590. SPECfp92 means that this is the report for the floating-point (FP) benchmarks in the 1992 release (the
earlier release was renamed SPEC89) The top part of the table describes the hardware and software. The bottom describes
the compiler and options used for the baseline measurements, which must use one compiler and one set of flags for all the
benchmarks in the same language. The tuning parameters and flags for the tuned SPEC92 performance are given in Figure
1.18 on page 49. Data from SPEC [1994].

We would like to think that if we could just agree on the programs, the experi-
mental environments, and the definition of faster, then misunderstandings would
be avoided, leaving the networks free for scholarly discourse. Unfortunately,
that’s not the reality. Once we agree on the basics, battles are then fought over
what is the fair way to summarize relative performance of a collection of pro-
grams. For example, two articles on summarizing performance in the same jour-
nal took opposing points of view. Figure 1.11, taken from one of the articles, is an
example of the confusion that can arise.

Computer A Computer B Computer C

Program P1 (secs) 1 10 20
Program P2 (secs) 1000 100 20
Total time (secs) 1001 110 40
FIGURE 1.11 Execution times of two programs on three machines. Data from Figure I
of Smith [1988].
Exploring the Variety of Random
Documents with Different Content
 In the act of bawling an order forward, Ensign Dave Darrin felt
his tongue hit the roof of his mouth. For, at this instant, the pursuing
canoe ranged up alongside the first.
There was a dim flash of something, accompanied by a yell of
unearthly terror.
“Light!” shouted Dave Darrin huskily.
“Aye, aye, sir.”
In a twinkling, the narrow, dazzling beam of one of the forward
searchlights shot over the water.
Within three seconds it had picked up the smaller of the canoes.
To the watchers from the deck of the gunboat this canoe appeared
to be empty.
Then the light shifted enough to pick up the second, larger
canoe, now darting shoreward under the impetus of two powerful
paddlers.
“Ahoy, there, shorebound boat!” yelled Ensign Darrin lustily. “Lay
to and give an account of yourselves!”
The challenged canoe moved on so rapidly as to call for the
constant shifting of the searchlight’s beam.
“Lay to, there, or we fire!” bellowed Ensign Darrin over the
rippling waters of Manila Bay.
But the canoe made no sign of halting.
“Sentry!”
“Aye, aye, sir.”
“Take aim and hold it!”
“Aye, aye, sir.”
Then again Dave challenged.
“Shorebound boat, third challenge! Lay to, instantly!”
No attention being paid by the two paddlers, Ensign Darrin now
gave the sharp order:
“Fire!”
That bullet must have whistled uncomfortably close to the fleeing
craft, for on the instant both paddlers rose in the canoe.
“Fire!” commanded Ensign Dave, the second time.
At the sound of the marine’s shot both poised figures sprang
overboard from the canoe.
 “Shall I fire again, sir?” asked the marine, as the beam of the
searchlight continued to play upon the waters where the divers had
vanished.
“Not unless you see those men that jumped overboard from that
canoe,” replied Ensign Darrin.
Though the searchlight continued to flash further across the
water, nothing was seen of the men from the canoe. Indeed, at the
distance, the rippling waves might easily conceal a swimmer.
“Pass the word for the boatswain’s mate!” Darrin ordered.
As that petty officer appeared, Darrin ordered him to turn out a
boat crew and put one of the boats over the side.
“First investigate the nearer canoe, then the second. Bring them
both in alongside. If you see any swimmers in the water, pursue and
pick them up.”
“Aye, aye, sir.”
Still the searchlight continued to play over the waters. The
“Castoga’s” small boat ranged alongside the smaller outrigger canoe,
and soon had it in tow with a line astern. A minute or two after the
second canoe was picked up. A short search was made for
swimmers, after which, on signal, the boatswain’s mate turned and
headed for the gunboat.
“Ship’s boat ahoy!” Dave called, as the boat and its tows came
near.
“Ahoy the deck, sir!”
“Are both canoes empty?” Darrin inquired.
“The first one isn’t, sir,” replied the boatswain’s mate. “There’s a
dead Chinaman in it. Head almost cut off; sword work, I should say,
sir.”
“Bring both tows alongside,” Dave ordered, with a shiver. “I will
communicate with the police.”
After ordering a wireless operator turned out, Ensign Darrin went
over the side, down a sea-ladder, to the smaller of the outrigger
canoes.
Huddled in a heap in the canoe, was a Chinaman who did not
seem to be more than thirty years of age. His head, nearly severed
from his body, had fallen forward until it hung close to the dead
man’s chest. It was only by turning the head that Ensign Darrin was
able to see the face, on which there still lingered a look of terror.
“A Chinese tong-fight or a gang murder,” Dave told himself, in
keen disgust.
Then climbing up over the side he sent an orderly to summon the
executive officer.
Less than three minutes later Lieutenant Warden, fully dressed,
and wearing his sword, walked briskly out upon the quarter-deck.
The executive officer listened intently while Ensign Darrin made
his report with conciseness.
“I’ll take a look at the body,” said Mr. Warden, and went down
over the side. He came up again, horror written in every line of his
face.
“A cowardly killing, Ensign Darrin,” declared the executive officer.
“Notify the Manila police by wireless.”
“Aye, aye, sir.”
“Call me again, if I am needed.”
“Aye, aye, sir.”
The instant Darrin had saluted and Mr. Warden had turned on his
heel, Dave, under a light just inside the superstructure, wrote a few
words which he signed in his official capacity as officer of the deck.
This was sent forward to the waiting wireless operator, who sent the
message to a military station on shore, whence the message was
telephoned to police headquarters.
Within three minutes the wireless operator, ran aft, saluting, and
reported:
“A police launch will put off immediately, and come out, sir.”
Fifteen minutes later a motor launch, flying the police ensign,
ranged up alongside the “Castoga.” An American official,
accompanied by four Filipino policemen, came on board.
Dave at once narrated what had happened, after which the
American police official inspected both canoes and looked at the
huddled yellow body.
“This will require investigation, sir,” declared the police official. “I
shall tow both canoes ashore, and then the force will get busy.”
 “Don’t you wish to send a wireless ashore, urging the police to
look out for two swimmers who are likely to attempt to land?”
suggested Dave.
“An excellent idea,” replied the police official, and wrote out a
despatch which Ensign Darrin sent to the wireless operator forward.
After that the launch chugged away with the two canoes in tow.
Twenty minutes later a wireless message was received aboard
the “Castoga,” and immediately the operator brought it aft.
“Native Policeman Rafeta,” Ensign Darrin read, “reports that a
Chinese swimmer was observed, by him, to land. The Chinaman
reported that his skiff had upset. Native policeman, not being
suspicious, reports that he allowed swimmer to proceed on his way.
Swimmer is to be identified by a fire-mark on the right cheek under
eye.”
“Burnt-face!” gasped Ensign Dave, recoiling slightly. “Then it
seems that I was not quit of that fellow when I turned my back on
him on the Escolta this afternoon. In what fiendish business can
‘Burnt-face’ be engaged?”
 CHAPTER III—MR. PEMBROKE BREAKS IN
On the next day the Manilla police had little of interest to add to the
account of the night tragedy on Manilla Bay.
Searching the city, and especially the Chinese quarters, the police
had been unable to find any yellow man answering to the
description of “Burnt-face.”
Very likely many of the Chinese residents of the city knew the
man who was sought, but Chinamen habitually mind their own
business, even to the extent of withholding important information
from the police. So within two or three days the chase was all but
forgotten. The Chinese “tongs” are secret societies that commit
killings in all parts of the globe where their people are to be found,
and the death of an unknown Chinaman does not provoke the police
anywhere to any great zeal in finding the slayer.
Then the “Castoga,” which, for reasons known only to the higher
naval authorities, had been anchored half a mile from the mouth of
the Pasig, was ordered to new anchorage off the naval station at
Cavite.
On board, the officers had ceased to speak of the strange
Chinese tragedy of the night; Dave and Dan had well-nigh ceased to
think about it.
One afternoon the French gunboat “Revanche” received visitors.
Ensigns Darrin, Dalzell and Hale were requested to represent the
“Castoga” and did so, going over in the launch.
On board the French boat they found a sprinkling of English and
Japanese naval officers. There were also a few officers from the
United States Army.
Our American friends were introduced to all present whom they
had not previously known. Half an hour later Darrin was inspecting
the “Revanche’s” lifeboat equipment under the escort of Lieutenant
Brun, of the French Navy, when a superior officer appeared on deck.
It was the same officer who had appeared, on shore, to exhibit such
marked disapproval of Mr. Pembroke.
“There’s an officer over there to whom I wish you would
introduce me,” Dave said to the lieutenant.
“With great pleasure,” replied Brun, “as soon as our turn comes.
That is Commander Bertrand, commanding the ‘Revanche.’ All the
gentlemen present will be introduced to him now.”
“If you don’t mind,” Dave added, quickly, in French, “I shall be
glad to wait until the last, as I should like to have a few words with
your commander.”
A group had gathered around Commander Bertrand, who, all
smiles and good will, played the host to perfection.
At last Lieutenant Brun led Dave over to be introduced. The
introduction accomplished, Brun moved away a short distance.
After the first few polite exchanges had been made on both
sides, Dave asked:
“Would you object, sir, to telling me whether you know a Mr.
Pembroke, an Englishman?”
“I know that it is a well-known English name,” replied
Commander Bertrand, “but personally I know no Englishman of that
name.”
“Do you remember seeing Mr. Dalzell and myself with a man in
front of the office of the Captain of the Port a few days ago?”
“I recall having passed you,” replied the Frenchman readily.
“That was Mr. Pembroke with whom we were talking.”
“Was it?” inquired the Frenchman politely, as he raised his
eyebrows. “Then perhaps I was in error. I felt that I had seen the
man before, but at that time his name was Rogers.”
“May I inquire, sir, if you know this man Rogers?”
Commander Bertrand shrugged his shoulders slightly as he
asked:
“Is he a friend of yours, Monsieur Darrin?”
“No; but he had presented himself to Mr. Dalzell and me, and
then had offered to do us a service.”
“I do not believe that I would trust him,” replied the Frenchman.
“I cannot say, positively, that Monsieur Rogers and Monsieur
Pembroke are one and the same man, but this I can assure you—
that Monsieur Rogers is far from being an honest man.”
Further than that the French officer seemed disinclined to discuss
the subject. After a brief chat on other topics Dave thanked the
French Commander courteously and moved away. In less than two
minutes, however, Dave found a chance to impart this information
briefly to Danny Grin.
“Pembroke looks like a good one to dodge,” declared Ensign
Dalzell.
“I don’t know,” returned Dave Darrin. “It all hinges on whether
he is really the chap who once called himself Rogers. Commander
Bertrand declined to be positive that they are one and the same,
though for himself, he seems to believe it. However, we are not likely
to see Pembroke again. He has made no effort to force himself upon
us.”
Not long after that the launch called, and the “Castoga’s” visiting
officers started to return to their own craft.
“There is some one waving to us,” declared Dave, staring across
the water at the occupants of a small motor boat.
“Why, it looks like Captain Chapin,” returned Dalzell.
“It is Chapin, and that is his sister with him,” returned Dave.
“See, she is standing up in the bow to wave her handkerchief to us.”
“Chapin ought not to allow her to stand up in the bow of such a
narrow craft,” said Danny Grin. “It’s a risky pose for any one but a
veteran sailor. It’s dangerous. She—”
“By Jove!” burst from Darrin. “There she goes—overboard!”
For a rolling wave, catching the small motor boat under the bow,
had rocked the little craft.
Miss Chapin was seen to stagger wildly and then plunge
overboard.
“They’ve stopped!” cried Dan. “She doesn’t come up, either!”
“Boatswain’s mate!” rang out Ensign Darrin’s voice sharply to the
naval launch alongside. “Put over there at once. Run astern of the
motor boat’s position.”
“Aye, aye, sir,” and the naval launch swung briskly around.
 “I beg your pardon, Hale, for forgetting that you are ranking
officer here,” Dave apologized, keeping his gaze out over the water.
“There’s no apology needed,” returned Ensign Hale. “Our only
need is to reach the spot as quickly as possible.”
The motor boat had stopped. Captain Chapin at the first
realization of the incident, had leaped up, and now stood scanning
the water for the first glimpse of his sister when she would rise to
the surface.
So great was the excitement on the naval launch that neither
Dave nor Dan really noticed it when another man aboard the motor
boat rose more slowly, showing his head for the first time above the
gunwale.
As the motor boat put about on her course both Captain Chapin
and this other man dived overboard.
“I wonder if they see Miss Chapin yet?” muttered Dave, as the
naval launch raced to the scene.
It was speedily apparent that Miss Chapin had not yet been
found, for both hatless swimmers swam about uncertainly, going
down head first, from time to time, as though to explore the water
near the surface.
Then the naval launch plunged into the scene. From it dived
three ensigns and two sailors aboard who were not engaged with
the handling of the craft.
With seven expert swimmers now in the water, Miss Lucy Chapin
stood an excellent chance of being found.
Hardly had the Navy men dived when Captain Chapin’s male
companion swam with long overhand strokes away from the rest.
“I see her!” shouted this swimmer, and dived.
“He has her!” panted Dalzell. “Hooray!”
Instantly six swimmers turned and swam toward the rescuer,
who now appeared on the surface supporting a woman’s head on his
shoulder.
“Good work! Fine!” cheered Dave.
Captain Chapin was the first to reach his sister’s rescuer.
“Is Lucy dead?” cried Chapin anxiously, when he beheld his
sister’s white face.
 “Stunned,” replied the rescuer. “I think she must have been
struck on the head by the boat as it passed her.”
Silently the other swimmers turned in behind the young woman,
her rescuer and brother.
“Better bring Miss Chapin to the ‘Castoga’s’ boat, Captain,” Dave
called. “It’s larger. We’ll take her directly to the gunboat and have
the surgeon attend her.”
The boatswain’s mate ran the naval launch up within easy
distance, and Miss Chapin was lifted aboard.
On one of the cushions Miss Chapin was laid, while all gathered
about her anxiously.
“Make the ‘Castoga’ with all speed,” ordered Ensign Hale. “The
young lady must have prompt attention.”
On the way to the “Castoga” Captain Chapin did everything he
could think of to revive his sister. The others stood about, ready to
help.
Then it was that Dave happened, for the first time, to face the
rescuer.
“Pembroke!” he called in astonishment.
“Howdy do?” asked the Englishman, with a smile holding out his
hand.
Though Dave felt himself chilling with suspicion of the pleasant
stranger, he could not withhold his hand.
“I was on my way out to visit your ship,” smiled Pembroke, as he
released Dave’s hand after a warm grip. “Captain Chapin was good
enough to say that he would present me on board.”
“And glad indeed I am that I undertook to do so,” exclaimed
Chapin. “If it hadn’t been for you, Pembroke, I am afraid my sister
would have been lost.”
Pembroke was now engaged in shaking hands with Dalzell, who
felt obliged to present him to Ensign Hale.
“A splendid rescue, that,” said Hale warmly.
The gunboat’s launch was now speedily alongside the “Castoga,”
the motor boat, a small craft that carried passengers on the bay for
hire, following at slower speed.
 “We’ve a half-drowned young lady on board, who needs the
surgeon’s attention,” called Hale, between his hands, just before the
launch ran alongside.
Miss Chapin was immediately taken on board, and carried to the
quarters of the executive officer, where she was laid in a bunk. Only
her brother and the surgeon remained with her.
Dave felt obliged to introduce Pembroke to his brother officers.
The Englishman proceeded to make their acquaintance with evident
delight.
Five minutes later the executive officer recovered his presence of
mind sufficiently to send ashore to Cavite for dry garments of a size
suitable for Miss Chapin’s use. In an hour or two that young lady,
revived and attired in dry clothing, was brought on deck on her
brother’s arm. She was weak, but out of danger.
“We came out in order to make a call aboard,” Captain Chapin
explained to the officers under the quarter-deck awning, “but we
had no idea we were going to make such a sensational visit.”
“I fancy that women are always nuisances aboard naval craft,”
smiled Miss Chapin, whereupon the assembled officers promptly
assured her that women were nothing of the sort.
In the meantime the three officers who had leaped over into the
bay had had time to change their clothing. It became a merry party
on deck.
Up to Mr. Warden stepped a messenger, saluting.
“The Lieutenant Commander’s compliments, sir, and will the
executive officer report to the Lieutenant Commander at once?”
“Immediately,” replied Lieutenant Warden, returning the salute,
taking his brief adieu by merely raising his uniform cap before he left
the party.
Ten minutes later Lieutenant Warden stepped briskly on deck. He
paused long enough to say something in an undertone to the officer
of the deck, who smartly passed the word for a messenger.
“I am sorry to announce,” said the executive officer, approaching
the group of officers who surrounded Miss Chapin, “that our
pleasant days in Manila are ended for the present.”
 “I should say so,” cried Captain Chapin. “There goes your recall
flag to the mast-head.”
“Right!” replied Mr. Warden crisply. “Our sailing orders have just
been wirelessed from shore. We sail at seven this evening, if our few
men on shore leave can be recalled in that time. Mr. Hale, you are to
take the launch and go ashore after the leave men.”
“Very good, sir,” replied that ensign, saluting, next raising his cap
to Miss Chapin and hastening away.
“Leaving, are you?” asked Pembroke, in a tone of regret. “And
what is your destination?”
“China,” rejoined Lieutenant Warden tersely.
The Englishman’s face changed expression.
“Not—” he stammered. “Not the—”
“For the Nung-kiang River,” replied the executive officer.
Dave Darrin and Dan Dalzell were the only ones present who
caught the strange, fleeting look that passed over the face of
Pembroke.
“Why can this Englishman object to our going to the Nung-kiang
River?” Ensign Darrin wondered. “What interest can he take in any
mission of ours there?”
 CHAPTER IV—THE LANDING PARTY AT NU-PING
“That ought to hold the pirates for a little while,” declared Danny
Grin, his good-natured face looking unusually grim.
“I think it will,” replied Dave, halting before his cabin door. “Dan
Dalzell, if my face is as dirty as yours I shouldn’t care to walk up
Main Street in my native town.”
“Go in and look at yourself,” scoffed Dalzell.
“It’s fully as dirty,” called Dave, from the interior of his cabin,
surveying himself in the glass.
But it was as honorable dirt as any man may have on his face—
the grime of powder-smoke as it blew back when the gunboat’s five-
inch guns had been swung open at the breech.
For the “Castoga,” intercepted by wireless on the way to the
Nung-kiang, had been sent to Hong Kong by an official order from
Washington. The threatened troubles along the Nung-kiang had
quieted down to such an extent that cautious officials in Washington
dreaded lest Chinese sensibilities should be wounded by the sending
of a gunboat up the river.
So, day after day, the “Castoga” had lain in the mountain-
bordered harbor at Hong Kong.
Then came the word one day that the Chinese rebels in the
district around the city of Nu-ping, on the Nung-kiang River, had
again become troublesome, and that the American mission buildings
at Nu-ping were threatened. The “Castoga” had been ordered to
proceed at full speed, she being the nearest craft of a draft light
enough to ascend the river.
During the last hours of darkness the gunboat had steamed up
the river, all eyes on board turned toward the sinister red glow that
lighted the sky above the Chinese city, capital of a province.
Just before daylight the gunboat dropped anchor with every man
and officer at quarters.
 From shore came the sound of rifle shots, a wild pandemonium
of yells, as thousands of raging Chinese surged upon the mission
buildings, to which fire had already been set, and from which the
American missionaries and their families, aided by the white
residents of Nu-ping, were making the only resistance that lay within
their power.
The first note of cheer that came to the missionaries and their
friends was the whistle of the gunboat, sounding clearly when still
two miles distant. Then the lights of the fighting craft came into
sight.
For a few minutes after coming to anchor, the commander of the
“Castoga” was forced to wait for sufficient daylight to enable him to
distinguish accurately between friend and foe.
At the side of the gunboat a launch and four cutters waited, to
carry a landing party, if the sending ashore of men should prove to
be necessary. Anxiously, using his night glasses every minute, the
American commander paced the deck and listened.
Then, when there was barely enough light, word was telephoned
to the division officers to begin action.
Boom! spoke the first gun from the gunboat. Other shots
followed rapidly.
In the compound before the burning mission buildings was a
mass of yellow fiends, crowding, yelling and shooting. From the
windows of such portions of the burning buildings as were still
tenable American rifle fire was poured into the mob.
That first shell, landing among the yellow fiends, killed more than
twenty Mongols, wounded others, and drove the attackers out of the
compound.
Boom! Bang! Other shells flew through the air, clearing away the
rabble further back.
From the mission buildings, a quarter of a mile away, went up a
wild cheer of hope.
But the attacking rabble, despite the first shell fire, came back,
inviting further punishment.
Again the gunboat’s five-inch guns roared out. There was now
sufficient light to enable the American gunners to make out the
locations of the mob.
At least thirty shells were fired ere the rebels beat a retreat
beyond the confines of Nu-ping.
It was time to stop firing, for some of the American shells had set
fire to Chinese dwellings and business buildings.
On a low hill, a quarter of a mile away from the burning mission
buildings, flew the Chinese flag, flanked by the flag of the governor
of the province.
Watching this yamen, or palace, the American officers saw a
body of not more than a hundred soldiers issue suddenly from
behind the walls. Straight to the mission hurried these tardy fighting
men. Though late in acting, the Chinese governor was sending an
invitation to the endangered missionaries and their friends to share
the hospitality and protection of his yamen.
“He might have done that before,” muttered Dan Dalzell.
“If he has so few Chinese soldiers,” Dave explained, “he never
could have driven back the thousands of rebels. Our friend, the
governor, is cautious, surely, but plainly he is no fool.”
Once the bombardment had stopped, the various officers, except
one division officer, had been ordered to their quarters to clean up
and put on fresh uniforms, for the work of the day was by no means
finished.
So back to their quarters hurried the released division officers.
Dave Darrin quickly divested himself of his dungaree working
clothes, then stripped entirely, going under a shower bath. From this
he emerged and rubbed down, drew on fresh underclothing, a clean
shirt, and hastily completed his toilet.
At that instant there came a summons at the door, with an order
for Ensign Darrin to attire himself in khaki uniform. The same order
was delivered to Dan.
“Landing party work,” was the thought that leaped instantly into
the minds of both.
Nor were they disappointed. Into the launch, with several other
boats alongside, tumbled forty sailors and twelve marines, armed,
and with rapid-fire guns and ammunition. In one of the other boats
were additional cases of ammunition; in others were commissary
supplies.
Dave received his orders from Executive Officer Warden.
“You will go ashore, Ensign Darrin, and at all hazards reach our
fellow Americans. What you shall do on reaching them will depend
upon circumstances and upon instructions signaled to you from this
ship. Ensign Dalzell will accompany you as next in command. On
board we shall keep vigilant watch, and you may rely upon such
backing as our guns can give you in any emergency that may come
up.”
Dave saluted, with a hearty “Very good, sir,” but asked no
questions. None were necessary.
In another moment the landing party had been reinforced by a
petty officer and three men who were to bring the boats back to the
“Castoga.”
Casting off, the launch headed shoreward, towing the boats
astern.
Within three minutes, landing had been made at one of the
smaller docks.
“I don’t see any reception committee here to welcome us,”
muttered Ensign Dalzell.
“Probably all of the natives, who are curious by nature, are
watching the burning of the buildings that our shells set on fire,”
returned Ensign Darrin. “But I’m glad there’s no reception party
here, for undoubtedly it would be an armed committee.”
As soon as landing had been effected, however, a petty officer,
who was sent forward with three men, succeeded in routing out a
number of sturdy, sullen coolies, who had been hiding in a near-by
warehouse. These yellow men the petty officer marched back briskly,
the coolies being forced to pick up and carry the ammunition and
food supplies.
“See to it that these Chinese don’t try to run away with the stuff,”
Dave ordered tersely. “Keep them under close guard.”
“Aye, aye, sir.”
At the word from Darrin, Dalzell ordered the sailors to fall in and
lead the way in double file, the marines marching at the rear of the
little baggage train.
“Straight to the yamen!” commanded Darrin, as he gave Dan the
forward order, then fell back to keep an eye over the conduct of the
porters.
For the first block of the march through the narrow, foul-smelling
streets, the natives contented themselves with glancing sullenly out
at the handful of daring invaders. But a turn in the street brought
the American naval men in sight of an angry-looking crowd of nearly
a thousand Chinese—all men.
“Are they going to block our way?” whispered Dan, marching
quietly on when Dave hastened to his side.
“They are not,” Darrin answered bluntly, “though they may try to.
No one is going to block us to-day until we have used all our
ammunition.”
“That has the good old Yankee sound,” grinned Dalzell.
Seeing that the sullen crowd was massing, Ensign Darrin went
forward, hastening in advance of his little column.
“Is there any one here who speaks English?” Dave called
pleasantly, above the dead hush of that stolid Chinese crowd.
There was no answer.
“All right then,” smiled Ensign Darrin, “I shall have to talk to you
by sign language. Make way, please!”
Drawing his sword, he signed to the Chinese to make way for his
command to pass. Still no response.
Ensign Dan, marching his men on, came up to Dave’s side.
“Column halt!” Ensign Darrin called promptly. “Order arms. Draw
bayonets. Fix bayonets!”
With a rattling of steel, accompanied by many grins, sailors and
marines alike obeyed.
“Once more, I call upon you to make way!” called Dave, striding
forward and endeavoring to wave the crowd aside by gestures with
his sword. Still nobody moved.
“Ensign Dalzell,” rapped out the sharp order, “form two platoons
extending across the street in close order. Give promptly the order to
charge.”
 As he gave this command Darrin stepped back, placing himself at
the extreme right of the first short platoon.
“Charge bayonets!” ordered Dan.
Dave led the men forward, Dalzell remaining behind with the
remainder of the little command.
Finding the points of the bayonets at their breasts, the Chinese
gave utterance to cries of fright. There was a backward surge.
“Halt!” cried Dave, just in time to prevent some of the Chinese
from feeling cold American steel. “Steady! Forward march! Hep, hep,
hep!”
Emphasizing the speed of the step with his “hep, hep,” Dave now
continued his squad at a brisk walk, giving the yellow natives time to
make their retreat without trampling one another.
At the next corner the Chinese surged off at right angles in two
directions.
“I guess we’ll find the rest of the way clear enough,” smiled
Ensign Dave, again forming his men in double file and falling back to
Dan’s side. “The Mongols had me scared. I was afraid I’d have to
order the men to load and fire.”
“Would you have done that?” asked Dalzell.
“Why not?” asked Dave, with a shrug of his shoulders. “There are
American women up at the yamen, and they are still in peril. My
orders are to reach the yamen, and I propose to do it if it be
possible. If any yellow men try to block our way they will do so at
their own risk. I’ll charge or fire into any crowd or force that blocks
our way.”
“Good!” chuckled Ensign Dan. “I like the sound of that talk!”
Down by the river front, save for the warehouses, the buildings
were of the meanest—flimsy affairs of bamboo, with cheaply
lacquered facings, windows of oiled paper and floors of earth. Now,
however, the little naval column began to pass through a better part
of the city. Here the houses were of wood, substantially built, and of
pagoda or tent patterns. Not a few of these dwellings were
surrounded by compounds, or yards, enclosed by high stone walls.
And then, at last, in the heart of the city, the column came out
upon the low hillside on which was the great square surrounding the
governor’s yamen.
None in front opposed Darrin’s command, but a crowd that must
have numbered two thousand followed close at the heels of the
detachment.
“Going to halt in the center of the square?” Dan inquired in a low
tone.
“No,” rejoined Ensign Dave. “I shall march up to the main gate in
the compound wall.”
“And then—?” inquired Dalzell.
“I shall demand to be admitted to the American refugees.”
“And if you are refused?” pressed Dan.
“That will be the governor’s worry,” replied Dave quietly.
 CHAPTER V—SIN FOO HAS HIS DOUBTS
It was a gray stone wall, some twelve feet high, that surrounded the
compound of the yamen. Sentries in the uniform of Chinese soldiery
were pacing the top of this formidable rampart.
Over the walls could be seen the strange, gracefully arched red
and yellow roofs of the several large and the few small buildings of
the yamen.
Under the gray walls, on the outside, crouched a few mangy-
looking beggars. Men and women of this type always loiter outside
of every yamen, trusting to the occasional generosity of the high
official who resides within, for in China every mandarin, governor
and other high official must always be a good deal of an alms-giver.
Not even the sight of the heavily armed little American column
stirred these beggars beyond the most ordinary exhibition of
curiosity.
“Put the column to the right oblique, and go over to that gate,”
directed Dave, pointing with his drawn sword.
A moment later the command, “Halt!” rang out. From the
ramparts above three Chinese soldiers gazed down stolidly.
Striding forward to the gong that hung before the gate, Ensign
Darrin struck it loudly three times.
A minute passed without answer. Dave sounded thrice again.
Another minute passed.
“Confound those fellows inside,” muttered Dave to his chum.
“I’ve heard, before this, that the Chinese official tries to show his
contempt for western barbarians by making them await his
pleasure.”
Glancing down his line, Darrin noted a sailor who was well known
for his physical powers.
“Henshaw!” summoned Dave crisply.
 Leaving the ranks, Seaman Henshaw stepped briskly forward,
saluting respectfully.
“Henshaw, do you think you could play a loud tune on this
gong?”
“Aye, aye, sir.”
“How long do you think you could keep that tune going?”
“An hour, anyway, sir.”
“Can you play that gong like a bass-drum?”
“Like a whole drum-corps, sir,” answered Seaman Henshaw, with
just the suspicion of a grin.
“Then fall to, Henshaw.”
Picking up the fancifully carved stick, Seaman Henshaw faced the
three-foot gong.
Bang! crash! zim! zoum! smash! It was a lusty tattoo that
Seaman Henshaw beat upon that resounding metal. The noise could
have been heard a mile away. Dave afterwards learned that every
sound was distinctly heard on board the gunboat.
 It Could Have Been Heard a Mile Away.

Bim! bam! whang! After a full minute of it Seaman Henshaw

looked as if he were still enjoying his task. Several of the men in the
waiting column had grounded their rifle butts that they might hold at
least one hand to an ear to shut out the din.
On the wall overhead the Chinese sentries moved uneasily away
from close quarters.
Crash! zam! bing! That gong rang forth as, it is safe to say, it had
not done before in centuries, for Henshaw was a young giant and
proud of his muscle and endurance.
Zim! zim! zam! The racket was more than ears could endure for a
long stretch.
 At about the end of the third minute the double gates were
thrown suddenly open. In the open gateway stood at least a score of
armed soldiers, at their head a young Chinaman, tall, well-dressed
and of rather commanding appearance.
Instantly Ensign Darrin pressed Henshaw’s unemployed arm.
With a final crash the pounding of the gong died out.
“His excellency, the governor, demands to know why this din is
being made at this gate,” declared the tall young Chinaman, in a
haughty voice, but in excellent English.
“It is my way of announcing my call,” Dave replied.
“Who are you?”
“Ensign Darrin, United States Navy, very much at your service,
sir,” Dave replied. “And now, sir, I have the honor to request that
you, also, announce your name and position here.”
“Since I serve his excellency, the governor, that is enough for you
to know,” replied the Chinaman. “However, I will state that my name
is Sin Foo. I am under secretary to his excellency, and, as such, I
have come to bring word to you that it is his excellency’s pleasure
that you depart from this neighborhood and return to your vessel.”
“I am very sorry, Mr. Sin Foo,” Dave responded, “but it is
impossible for me to make my call on a proxy. I must see his
excellency in person.”
“I am very sorry,” replied the secretary, speaking in a tone of cold
contempt, “but his excellency cannot see you so early in the day.
Later—”
“Attention,” called Ensign Dave, in a low voice. “Column, forward
march!”
Instantly the naval line moved forward. Shocked and indignant,
the secretary spoke in Chinese to some of the soldiers. The big gate
began to move as though it would shut.
“At the double quick! Charge!” shouted Dave Darrin, leaping
forward, brandishing his sword.
In a twinkling the first dozen seamen, headed by two officers,
had rushed into the compound.
At one side stood two Chinese soldiers, working a cumbersome
wheel, attached to a windlass and rope that moved the double gate.
Henshaw leaped at this pair, knocking both down.
“I must warn you, Ensign Darrin,” shouted Sin Foo, his face
purple with rage, “that this conduct of yours is contrary to the
usages of respect that must be observed between the
representatives of two great countries. Your conduct, sir, is an
outrage!”
“And the governor’s conduct also is an outrage,” Darrin retorted
sternly, “in allowing mobs to burn the mission buildings and all but
take the lives of the American missionaries and their families.”
“All the mission Americans are safe at this yamen,” retorted Sin
Foo. “There is no need to fear for your fellow Americans. They are
safe and under the immediate protection of his excellency, the
governor.”
“That is what I have come to see about,” Dave declared. “Mr. Sin
Foo, I have no wish to be lacking in courtesy, and I shall display as
much as I can, under the circumstances. But my men are now inside
the compound, and here they will remain until my orders are
changed by my commanding officer.”
Though the Chinese soldiers had withdrawn to varying distances
before the harmless bayonet rush, Sin Foo remained and faced
Ensign Darrin with every sign of indignant disapproval in his almond-
shaped eyes.
The impressed Chinese porters had been driven into the
compound, where they dropped their burdens. Dan quietly paid
them off with silver coins. The instant they found themselves
permitted to leave, these sullen coolies fairly flew out through the
still open gate.
“Sir,” began Sin Foo again, speaking with great haughtiness, “if
these supplies are intended for the American missionaries, I will
undertake to receive them on behalf of your countrymen, but I must
once more, and for the last time, insist that you withdraw your men
from this compound. If you do not instantly withdraw, it will provoke
grave trouble between your country and mine.”
“Mr. Sin Foo,” Dave replied, speaking pleasantly, and smiling, “I
wish to treat you, and all other Chinese officials with every mark of
courtesy. I must make it plain, however, that I shall not leave this
yamen until I have been so ordered by my commanding officer.
Moreover, I am under strict orders to see the American mission party
at once, and I must very respectfully insist upon no more delay. I
demand, sir, to see Bishop Whitlock first of all.”
“Ensign Darrin, your language is insolent!” cried Sin Foo angrily.
“My talk will quickly change to acts, if my requests are not at
once granted,” replied Dave, firmly.
“‘Acts’? What do you mean by that word?” demanded Sin Foo.
“I mean that if Bishop Whitlock and his friends are not at once
produced, I shall be under the necessity of searching the yamen for
them,” Ensign Darrin rejoined.
“Search the yamen?” gasped Sin Foo incredulously. “Would you
dare profane the sovereignty of China?”
“I’ll do it in just five minutes, if my request is not heeded,”
retorted Ensign Dave drawing out his watch.
By this time at least one hundred and fifty armed Chinese
soldiers had appeared, on the ramparts, in the compound, in the
doorways and windows of the buildings. Darrin’s force was much
inferior numerically. Sudden treachery on the part of the Chinese
might cut the American naval force in two, but Danny Grin was
keeping alert watch on all Chinese in sight.
“You are making a grave mis—” began Sin Foo loftily.
“And you have already lost forty-five seconds of that five
minutes,” Ensign Dave suggested, still standing, watch in hand. “If
you use up the time in conversation, Mr. Sin Foo, I shall not grant a
grace of even five seconds.”
“Your insolence, sir, overwhelms me,” replied the under secretary.
“Shudderingly, I shall take it into the presence of his excellency.”
“And impress upon his excellency, if you please, that I am not
going to lose time,” answered Dave, again glancing at his watch.
Turning on his heel, Sin Foo disappeared through a near-by door
of one of the buildings.
Several minutes slipped by. Dave glanced frequently at the hands
of his watch.
“The time is nearly up, Dan,” he announced, at last. “You remain
in command of the marines and guard our ammunition and other
stores. At the second of five minutes I shall form the sailormen and
march through this yamen until I find the missionary party.”
Danny Grin nodded gravely.
“Seamen fall in!” called Ensign Darrin, replacing his watch in his
pocket. “Forward, guard left, mar—”
“Stop!” cried a ringing voice. Out of the doorway through which
he had vanished appeared Sin Foo, running and waving his arms.
“The governor’s answer?” Dave curtly demanded, turning upon
his heel.
“I will take you to the missionary party,” conceded Sin Foo.
“Very well; step with me, then, and lead the way.”
“But you must not take an armed party with you,” protested Sin
Foo, looking very much aghast.
“My men go with me,” Dave replied firmly. “Sir, we cannot have
any more nonsense. I am convinced that my countrymen must be
prisoners, else they would have come out to greet me before this.
Lead the way and I will march my men behind you.”
Looking as though he would very much like to say a good deal,
Sin Foo led the way around the buildings to the left, thence to the
farthest building of all at the rear of the compound. Scattered
around the outside of this building were nearly a score of Chinese
soldiers carrying their rifles at shoulder arms.
“You have kept the Americans as prisoners, just as I suspected,”
charged Ensign Darrin, turning accusingly upon the under secretary.
“And you forget, Ensign Darrin,” retorted Sin Foo, “that his
excellency the governor commands here.”
“We’ll let it go at that,” answered the young ensign, “provided
your governor doesn’t attempt to put any crimps in the safety or
liberty of my countrymen. Right now, be good enough to order your
soldiers away so that there may be no clash between them and my
men.”
Through the windows of the one-story building Dave Darrin could
see several faces of men and women looking eagerly out.
Sullenly, Sin Foo spoke to the Chinese soldiers, who, saluting,
withdrew to a distance, though they did not leave the scene. Then a
door was flung open, and American citizens began to pour out.
 Darting through the foremost of the throng was one handsome
young American woman, who, holding her arms outstretched, while
eager tears of gladness glistened in her eyes, cried:
“Dave!”
It was Belle Darrin, once Belle Meade, Dave’s schoolgirl
sweetheart then, and now his wife.
“You, Belle?” he exclaimed, almost incredulously, as he embraced
her. “I thought you were in Manila, awaiting word when and where
to join me.”
“I couldn’t wait any longer to join you, so I came up in the last
steamer from Manila, and transferred to a river boat at the foot of
the river. Aren’t you glad to see me?”
“Glad, indeed!” Dave embraced her again. But he was on duty,
and most urgent duty at that. Even further conversation with his
beloved wife must wait until he had rightful leisure.
Then his eye fell upon another in the little throng.
“You here, Mr. Pembroke?” Ensign Darrin inquired.
“Yes,” confessed the Englishman. “I’m a bit of a rover, you know.
Never know where I shall be next.”
“And Mr. Pembroke has been extremely kind in helping me on the
journey,” Belle added brightly. “Mr. Pembroke told me that he had
met you in Manila.”
Though Dave bowed courteously he couldn’t help feeling dislike
of this smooth-talking Pembroke. The latter was an Englishman;
then, unless he was serving his country in this part of the world, why
wasn’t the fellow at home, doing his bit of military service for
Britain? He was young enough, and able-bodied, and England was
calling all her younger men to the colors. To Darrin’s mind it was a
sheer confession of disgrace for Pembroke to admit that he was
merely touring the world at a time when England was demanding
service on the battle field from every young Briton who was not
otherwise engaged in serving his country.
“When you have time, Mr. Darrin, I’ll claim just a word of
greeting,” said a soft voice, and a gloved hand was held out to Dave.
“So you came through also, Miss Chapin?” Dave inquired, as he
took Lucy Chapin’s hand.
 “I’m glad to see you, but sorry you’re here,” rejoined Dave.
“Why sorry to see me here?” inquired Miss Chapin. “Aren’t we
now under the protection of the American Navy?”
“Every sailorman on the ‘Castoga’ will die willingly in defense of
this party,” Darrin told her, “but the trouble may easily assume such
proportions that our little force will prove wholly inadequate.”
Then, glancing swiftly over the missionary party, the young naval
officer added:
“Will some one kindly introduce me to Bishop Whitlock?”
As Dave had expected, it was one of the three white-haired men
of the party who now pressed forward. Mrs. Darrin introduced her
husband to the bishop.
“You reached us not a bit too early,” the bishop assured Dave.
“You were practically prisoners in the yamen, sir?” Dave asked.
“Almost, I fear, though we refused to give up our arms. Even now
seven of our men are inside keeping guard over our weapons.”
“How many rifles do you have?” Dave asked.
“Thirty-two,” answered the bishop promptly. “The American
residents of the city flocked to our defense.”
“From what I saw from the ship’s deck,” rejoined Darrin, “all I
can say is that you presented a magnificent front with only thirty-
two rifles. As I have but fifty-two naval rifles with me, that makes up
a total force of only eighty-four rifles.”
“Can’t we get through to the water-front?” inquired Belle. “For
you are going to take us to the ‘Castoga,’ are you not?”
“If we can safely get there,” Dave replied. “And now I must drop
everything else until I have communicated with the gunboat. Bishop,
did you lose any of your party?”
“None of the white members,” replied the missionary. “Our
sixteen Chinese converts at the mission insisted on taking care of
themselves. Whether any of them has been killed, I do not know.”
“I must get a signalman up on the walls,” Dave continued.
“Bishop, will you kindly see, sir, that your party follows my men? I
am going to the other side of the compound.”
As soon as Belle Darrin caught sight of her old school friend,
Danny Grin, she hurried forward to greet him.
 Out of the main building of the yamen came Sin Foo, with sullen,
offended face and stately tread.
“Sir,” called Dave, “I must put a signalman up on the ramparts.”
“Since you take everything into your own hands,” replied the
secretary coldly, “you do not need his excellency’s permission. Yet I
am charged to say that all you do here is against the protest of his
excellency, and complaint will be made to your government.”
“I am sorry, sir, to seem to show discourtesy,” Ensign Dave
replied, “but all that I do here is under general instructions from the
highest representative of my government in these waters.”
With that Dave called a signalman to him, gave him a message,
and directed six other sailors to climb, with the signalman, the inside
steps that led to the rampart.
No sooner had the signalman, in the lead, gained the rampart,
than a five-inch gun on the “Castoga” boomed out.
“Ensign Darrin, sir,” bawled down the signalman lustily, “I think
you will be glad to be up here, to see what is going on.”
Dropping Belle’s hand, which he had just taken, Ensign Dave
darted up the steps, uttering, on reaching the top of the stone wall,
an exclamation of dismay.
“Ensign Dalzell!” he shouted, beckoning the summons to his
brother officer.
 CHAPTER VI—HECKLING HIS EXCELLENCY
“Jupiter!” gasped Dan, as he reached Dave’s side.
Boom! bang! Two shots were fired almost together from the
“Castoga’s” forward guns.
“The rebels are returning from the suburbs,” Dave exclaimed,
“and even the near-by houses are emptying themselves of hundreds
of other armed men.”
“There must be a million of them, in all,” said Danny Grin briefly,
“but I reckon we can thrash ’em all.”
“We’ll have to, or go under,” was Dave’s brief retort. “There
cannot be a doubt that the armed multitude intends to attack the
yamen.”
In the meantime Signalman Ross was sending the message that
Dave had given him. Now a signalman on the gunboat wig-wagged
back:
“Do not attempt to leave yamen with your party until you receive
orders so to do.”
“I’m glad of that command,” Dave muttered to his subordinate. “I
wouldn’t care to risk any of our American women by trying to take
them through such a rabble as I see advancing.”
Again some of the “Castoga’s” guns spoke. The shell fire was
doing some execution in the ranks of the oncoming rebels, though
not enough to halt their march.
“I am going down into the compound to send up men and rapid-
fire guns,” Dave announced to his chum. “Post the men, and station
one rapid-fire gun on each of the four sides of the compound.”
“What are you going to do with the Chinese soldiers?” Dan
asked.
Dave frowned.
“I don’t know,” he said. “This is the governor’s yamen, and these
are his troops. I don’t believe we can trust them, but, on the other
hand, have we any right to drive the soldiers out? And would they
go peaceably, or would they open fire and put the women in
danger?”
“Ask the Captain, by signal,” Dan advised.
“Ask him yourself, signing my name, Dan. Whatever we do, the
rapid-fire guns can’t be placed on these walls a moment too soon.”
Hustling below, Dave ordered up the sailors and all but four of
the marines. Each man, as he went, carried up a case of one
thousand cartridges, either for the rapid-fire guns or for the infantry
rifles.
“You men in charge of the stores,” Dave ordered, “keep the
Chinese away from our property. Don’t let any of the yellow men
touch the stores. Shoot before you permit that, and shoot promptly
and to kill!”
Then Darrin turned his attention to the missionary party. Of the
thirty-two men who carried rifles, he sent twenty to the ramparts,
while he directed the other twelve to stand guard over the women.
Having made these dispositions of his command, Ensign Darrin
again raced up to the top of the wall.
“There’s the answer just coming from the gunboat,” Dan
informed him. So Dave, shading his eyes with one hand, picked up
this message from the “Castoga”:
“Ensign Darrin: You will need to exercise great caution as to what
you do in the yamen, as only the presence of imperilled missionary
party justifies presence of your command there at all. Better consult
with the governor.”
“That’s just what I’ll do,” Dave uttered grimly. “That governor
chap has been keeping himself mighty well out of sight. Now it’s
time for me to see him, and he must show up and take some little
hand in affairs that are going on in his city and province.”
“Shall I allow more Chinese soldiers up here on the wall?” asked
Ensign Dalzell. “Here they come.”
That was, indeed, only too true. Very quietly, under their own
officers, some hundred and sixty of the governor’s troops had
formed in four detachments, going to the walls at the four sides of
the compound and starting up the steps.
 “I don’t know whether we can stop them, and I don’t know that
they won’t be willing to fight with us and for us,” returned Darrin,
perplexedly. “I’ll follow the commander’s orders and see the
governor at once.”
Running down, and darting across the compound, Dave halted
before the principal door of the main building, the door Sin Foo had
used.
Knocking lustily with the hilt of his sword, Dave did not wait more
than thirty seconds. Then reports from two more of the gunboat’s
guns decided him. He seized the latch, trying to force the door, but
only to find that barrier locked.
“Open!” ordered Dave, in his loudest quarterdeck voice. “Open!”
He waited another thirty seconds, but no one inside obeyed.
“Open,” he shouted, “or I shall order my men to batter the door
down!”
Inside, instantly, he heard the murmur of voices.
“Well,” demanded the irate young officer, “will you open, or do
you wish the door battered down?”
Preceded by a rattling of bolt chains, the great door was thrown
open. Into the doorway breach stepped Sin Foo, calmly disdainful.
Behind him stood fully a score of Chinese soldiers, each with rifle
leveled ready to shoot.
“Why this unseemly disturbance before the official residence of
the governor?” demanded Sin Foo.
“I must see the governor immediately,” Dave replied.
“It will be impossible to see his excellency, except upon
appointment,” replied the secretary. “His excellency’s presence is
sacred, and is not to be invaded at will by a hasty caller with sword
in hand. If you will wait here, I will ascertain if his excellency will be
pleased to see you in an hour.”
“If he keeps me waiting two minutes,” Ensign Dave retorted, “I
shall search this building for him.”
“At your first step inside,” Sin Foo proclaimed, “these soldiers will
fire upon you. That will be the signal for all our troops to fire on your
men, who are no better than unlawful invaders.”
“Ensign Dalzell!” shouted Dave, over his shoulder.
 “Aye, aye, sir.”
“Rush six men here, with the machine gun from the river side of
the wall!”
“Very good, sir!” came in Dan’s delighted voice.
No sooner did he comprehend than Sin Foo uttered something in
Chinese. Through the squad of soldiers darted half a dozen yellow
servants who instantly sought to close the door.
“Back with you!” ordered Darrin, whipping out his revolver and
menacing the frightened servants.
“If the word to start killing is given it will really come from you,
Mr. Sin Foo,” Dave warned the secretary sternly, “and you will be one
of the first men to drop dead.”
Dave’s foot was now posted where it would obstruct the closing
of the door, even if attempted.
“Here we are, sir,” panted a sailor, darting up with a machine gun
and its crew.
“Train your gun to cover this corridor,” Dave ordered, tersely.
The gun captain posted the machine gun so that its nose pointed
at the squad of Chinese.
“Withdraw those soldiers, Mr. Sin Foo!” briefly commanded
Ensign Darrin.
“What?” gasped the thunderstruck under secretary. “You
presume to give orders in the governor’s very residence?”
“I don’t intend to argue,” Dave retorted, as another of the
“Castoga’s” guns spoke from the river. “If you don’t run your soldiers
out of this corridor, then the janitor will have them to sweep out, for
I’m going to order the machine gun into action mighty soon!”
Sin Foo looked puzzled, but soon he spoke to the soldiers, who,
scowling, wheeled and marched back down the stone-flagged
corridor, vanishing around an angle of the wall.
“The governor will not see you, sir,” Sin Foo insisted.
“You’re wrong there, too,” Dave crisped out. “It was my wish to
be courteous. But now I have the honor to tell you that the governor
will come to the doorway to speak with me, and he’ll come very
promptly, or else I shall march a force of men into the house and
find him. It will be much pleasanter for his excellency if he promptly
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com