Core Performance
Core Performance
Presented by :
Narendra
Tools
• Pythonsv
• Harmony
• EDP(EMON data processing)
• EMON(Event monitor)
• Maple
• Latte
• Linux(red hat)
Why performance is important
1.To make Intel products competitive, performance is critical beside functional quality.
2.Since we don’t know all workloads that data center customer may use, we are using more proxy
workloads to optimize performance.
3.performance is not only dependent on hardware but software as well. Too many junk files on the
hard drive, corrupted memory, and even invalid registry keys can make a PC with adequate specs
perform far less than the sum of its parts.
4.For gaming and video editing, you should choose a good computer that has between four and eight
cores as the increased number.
5.For the best value when shopping for a new PC, you should be on the lookout for CPUs with
multithreading capability. Intel calls theirs Hyper-Threading and AMD calls theirs Simultaneous
Multithreading.
6. The CPU’s clock speed is another factor to consider. Clock speed is the rate at which the processor
operates, the storage and RAM affect the speed of the processor as well.
Multitasking
• In computing, multitasking is a method by which multiple tasks, also known as processes, share common
processing resources such as a CPU. With a multitasking OS, such as Windows XP, you can simultaneously run
multiple applications
Cache memory
• The cache memory of the CPU cache plays a very important role in the smooth and fast functioning of a
processor. It basically keeps the program instructions and commands which are recently used that the CPU
might need in the future.
• The CPU can retrieve this data at a faster rate from cache memory rather than taking it from the main
memory.
• The overall performance and speed of the processor improves with these fast retrievals and thus make your
device much smoother and faster.
• Cache memory is a chip-based computer component that makes retrieving data from the computer's
memory more efficient. It acts as a temporary storage area that the computer's processor can retrieve data
from easily. This temporary storage area, known as a cache.
• Cache is more accessible to the processor, and able to increase efficiency, because it's physically close to the
processor.
Types of cache memory
Cache memory is fast and expensive. Traditionally, it is categorized as "levels" that describe its
closeness and accessibility to the microprocessor. There are three general cache levels:
L1 cache, or primary cache, is extremely fast but relatively small, and is usually embedded in the
processor chip as CPU cache. The size of L1 cache varies from 2KB to 64KB.
L2 cache, or secondary cache, is often more capacious than L1. L2 cache may be embedded on the
CPU, or it can be on a separate chip and have a high-speed alternative system bus connecting the
cache and CPU. The size of L2 cache varies from 256KB to 512KB.
Level 3 (L3) cache is specialized memory developed to improve the performance of L1 and L2. L1 or
L2 can be significantly faster than L3, though L3 is usually double the speed of DRAM. Generally used
to boost up performance of L1 & L2 cache.
Cache memory is important because it improves the efficiency of data retrieval. It stores program
instructions and data that are used repeatedly in the operation of programs or information that the
CPU is likely to need next.
Memory Bandwidth
• The maximum memory bandwidth is the maximum rate at which data
can be read from or stored into a semiconductor memory by the
processor.
• Bandwidth can be measured bytes/second.
• memory bandwidth can be calculated from the memory clock and the
memory bus width.
• Memory bandwidth, depends on multiple factors, such as sequential or
random access pattern, read/write ratio, word size,
Latency
• Latency refers to the time the operation takes to complete.
• Latency is "the delay that the instruction generates in a dependency
chain", or the cycles it take from when the instruction starts executing
to the time you get the correct output. So there is nothing called
"latency of a CPU".
• Latency depends on factor cache memory as well as clock, workloads.
• Latency can be measured (ns).
• If you are streaming lot of data through the processor and get lot of
requests then you will have limited bandwidth and latency will be
reduces.
Threads
• A thread is a virtual version of a CPU core. To create a
thread, Intel CPUs uses hyper-threading, and AMD CPUs uses
simultaneous multithreading
• Multithreading is a technique which allows a CPU to execute many
tasks of one process at the same time. These threads can execute
individually while sharing their resources.
• Each CPU core can have two threads. So a processor with
two cores will have four threads. A processor with eight cores
will have 16 threads. A processor with 24 cores (yes, those
exist), will have 48 threads.
Types of core processor
• Single core processor
• Multi-core processor
In single core processor there are two main issues
• To execute the tasks faster, you need to increase the clock time.
• Increasing clock time increases power consumption and heat
dissipation to an extremely high level, which makes the processor
inefficient.
Simultaneous multithreading(SMT)
• Simultaneos multithreading is a technique for improving overall
performance of CPU.
• Intel branded this process as hyper-threading, but hyper-
threading is the same thing as simultaneous multithreading.
For example, AMD CPUs with four cores use simultaneous
multithreading to provide eight threads, and most Intel
CPUs with two cores use hyper-threading to provide four
threads.
Cache coherancy
• In a shared memory multiprocessor system with a separate cache memory for each processor, it is
possible to have many copies of shared data: one copy in the main memory and one in the local
cache of each processor that requested it. When one of the copies of data is changed, the other
copies must reflect that change. Cache coherence is the discipline which ensures that the changes
in the values of shared operands (data) are propagated throughout the system in a timely fashion.
Thank you