Terminology

Performance analysis is a quite comprehensive topic. It fills whole books and is surely out of scope of this book. What is attempted here is to give a short introduction into that subject. Following is a concise definition of the three most important concepts used in performance analysis literature:

Load

Throughput

Response Time

 

Load

A Web site - and especially the application that is running behind it - typically behaves and performs different depending on the current load, that is, the number of users that are concurrently using the Web site, at one point in time. This includes clients who actively perform requests at a time, but also clients who are currently reading a previously requested Web page. Peak load often refers to the maximum number of concurrent users using the site within some point in time.

 

Throughput

A Web site can only handle a specific number of requests in parallel. Throughput depends on that number and on the average time a request takes to process; it is measured in requests/second. If the site can handle 100 requests in parallel and the average request takes one second, the Web site's throughput is 100 requests per second.

 

Response Time

Response time refers to the timeframe from when the client initiates a request until it receives the response. Typically the time taken to display the response (usually the HTML data inside the browser window) is also accounted for in the response time.

  Prev | Home | Next

 

WebSphere is a trademark of the IBM Corporation in the United States, other countries, or both.

 

IBM is a trademark of the IBM Corporation in the United States, other countries, or both.