9.1.Measuring Performance

9.1.Measuring Performance - Web Applications Engineering:...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Web Applications Engineering: Measuring Performance Service Oriented Computing Group, CSE, UNSW Week 9 Material in these Lecture Notes is derived from: Performance by Design: Computer Capacity Planning by Example by D. Menasce, V.A.F. Almeida, L. Dowdy, Prentice Hall The Art of Computer System Performance Analysis by Raj Jain, Wiley Press S. Venugopal (CSE, UNSW) COMP9321, 11s1 Week 9 1 / 26 Planning for Performance Performance Engineering- Predicting the expected performance of a system and providing recommendations to realize the optimal performance level. Activities involved in performance engineering: 1 Understand the key factors that affect a systems performance 2 Measure the system and understand its workload 3 Develop a workload model that approximates the actual workload 4 Develop an analytical model that predicts performance of the system 5 Run the models under different scenarios to forecast system performance S. Venugopal (CSE, UNSW) COMP9321, 11s1 Week 9 2 / 26 Performance Metrics Response time- the time it takes a system to react to a human request. Also called Round-trip Time (RTT) RTT depends on the processing time at the client, the time taken for the request and the reply to travel through the network, and the time involved in processing the request at the server. A browser incurs at least 3 RTTs : DNS Name Resolution, TCP Connection Setup, and HTTP Request-Response. Throughput- The rate at which requests are completed from a computer system is called throughput Measured in operations per unit time - e.g. Requests per second, Sessions per second, Page views per second or transactions per second Under increasing load, throughput can reach a maximum and plateau or even decrease - thrashing S. Venugopal (CSE, UNSW) COMP9321, 11s1 Week 9 3 / 26 Performance Metrics Availability- fraction of time the system is up and available to its customers - - also called its uptime . Q: A system with 99.99% would be unavailable for how much time in 30 days ? Availability is affected by failures and overloads Availability is also a parameter in system design Reliability- is the probability that a system is going to function properly and continously in a fixed period of time Reliability over a large period of time tends to availability. Resource Utilization- The amount of resources consumed by an application, expressed as a percentage of resources available Resources can be of any type : CPU, storage, database, bandwidth, etc. Other non-functional requirements are: security and extensibility. S. Venugopal (CSE, UNSW) COMP9321, 11s1 Week 9 4 / 26 Performance Metrics Performance objectives are cast in terms of metrics.e.g. I 99.999% availability during Nov-Dec I Max. response time of 10 ms for the home page of a website I Min. throughput of 2 transactions per second....
View Full Document

Page1 / 33

9.1.Measuring Performance - Web Applications Engineering:...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online