Scribd
Upload
120 views

Performance Engineering

This document discusses software performance engineering. It defines software performance as the degree to which a software system meets objectives for timeliness, such as response time and throughput. Performance engineering is the process of planning and evaluating a system's performance throughout development to enhance responsiveness and usability while preserving quality. However, performance engineering is not widely practiced due to a cognitive gap between software development and performance analysis domains and a lack of resources prioritizing shorter development times over performance activities. Many performance problems stem from architectural issues requiring performance modeling early in the architecture phase.

Uploaded by

pramod_kmr73
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
120 views

Performance Engineering

This document discusses software performance engineering. It defines software performance as the degree to which a software system meets objectives for timeliness, such as response time and throughput. Performance engineering is the process of planning and evaluating a system's performance throughout development to enhance responsiveness and usability while preserving quality. However, performance engineering is not widely practiced due to a cognitive gap between software development and performance analysis domains and a lack of resources prioritizing shorter development times over performance activities. Many performance problems stem from architectural issues requiring performance modeling early in the architecture phase.

Uploaded by

pramod_kmr73
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 11

Performance Engineering

Software Performance
• The degree to which a software system or
component meets its objectives for
timeliness
– Time required to respond to specific events
(response time)
– Number of events processed in a given time
interval (throughput)
Software Performance
• It’s all about timing!
• Events occur
– Messages, interrupts, user request, clock tick
• The system must respond to the events
• Performance: How long does it take for the
system to respond to an event
– Parameters: number of events and their
patterns
Performance Scenario
• A request for some service arrives at the
system
• Satisfying the request consumes
resources
• During this time the system may be
simultaneously serving other requests
Software Performance
• High Availability Defined
• High Availability (HA) is a design and
implementation that ensures a certain
degree of operational continuity.
In other words…
• The system is up
• The users are happy
• The business is not losing money due to
outages
Basic Model
Parameters
• Customer population is infinite
• Arrival time of each customer
– Time interval between 2 customers
• Random variables with common distribution function (A)
– Service time
• Random variables with common distribution function (B)
• Number of servers
• Size of the waiting line (queue) – infinite or finite
• Service discipline
Dimensions of Software
Performance
• Responsiveness: the ability of a system to
meet its objectives for response time or
throughput
– User’s perspective vs. system’s perspective
• Both subjective and objective
• Scalability: ability of a system to continue
to meet its response time or throughput
objectives as the demand for the software
functions increases.
Software Performance Engineering
(SPE)
• Systematic process for planning and
evaluating a new system's performance
throughout the development life cycle
• Goals:
– Enhance the responsiveness
– Enhance the usability
while preserving the quality
Why SPE is not Widely Practiced?
• Cognitive gap between the software
development domain and the performance
analysis domain
– Software developers do not know much about
performance modeling
• Lack of resources: pressure for “shorter
time to market” leaves SPE out of the list
of must-do activities
Essential Difficulty
• Many performance problems are usually
architectural in nature
– Rooted at the design and architecture stage
• Performance Modeling must start at the
architecture phase
– Research: derive performance models from
design models (e.g., UML, ADLs)
• Structure
• Behavior

You might also like