CS457 - System Performance Evaluation - Winter 2008
Questions and Comments
- Office hours for A1.
Lecture 11
Measuring Performance
Traces
What I have called a log is usually called a trace, which is
- a time-ordered sequence of events from a real system, as opposed to a
simulation
- useful for assessing performance and generating workloads
Monitors
The code inserted to generate traces.
- often include capability to analyse and display results
Measurement
What other reasons might you have for wanting to measure performance?
- Identify heavily used components
- hardware or software modules
- Locate bottlenecks
- Characterize workloads
- Validation of simulations or models
What do we try to measure
- Arrival and departure times
- Processor activity, which can give service times
- Other resource activity
- Failures
Levels of Measurement
- Profiling tools
- These are plentiful and easy to use.
- They don't require any changes in application code
- How do they work?
- Application code
- Operating system
- Kernel
- Hardware
The Big Question
How much does the introduction of monitoring software influence the
performance of the application.
- Event-driven monitors
- Sampling monitors
Analysing Data
You have some data. What next?
- Can you see any patterns? Exploratory data analysis.
- Are the patterns real?
- How big are the patterns?
- Does it matter? Should you do anything about it?
- Consider the parameters in their technical context
Reprise the patterns we saw in the application of Little's Law.
Return to: