Scoring Methodology

WebMark 2004 is made up of 2 scenarios consisting of the following websites:

Information Processing

• Portal
• Research 
• Training 

Commercial Transactions

• Finance 
• Marketplace 
• Purchasing 

For a description of these scenarios see Scenario Description. Each of these scenarios consists of 3 websites each.

Website Rating

Each of the Websites above have a rating that is calculated by taking the sum of duration of operations that constitute that Website and then comparing it with the equivalent sum of the duration of operations on the Calibration System. The rating is calculated by dividing the sum of the duration of operations in the Website on the calibration system by the sum of the duration of operations on the test system and multiplying by 100. The result is then rounded to the nearest integer. At the end of a benchmark run, the Website rating for each of the 6 Websites is calculated.

Scenario Rating

The Scenario rating (for each of the 2 scenarios) is calculated by taking the Geometric mean of the Website ratings of all the three Websites within each scenario. The scenario rating is rounded to the nearest integer.

WebMark 2004 Overall Rating

The WebMark 2004 overall rating is calculated by taking the geometric mean of the two Scenario ratings. The WebMark 2004 rating is rounded to the nearest integer.  

Official Run

An official run of WebMark 2004 runs a trial run followed by three iterations of the benchmark. On Windows XP, the ProcessIdleTask API is called after the trial run in order to optimize Windows XP performance.  In an Official Run, the scoring methodology involves a few extra steps. The site Rating is calculated as above for each Website in each iteration. An average site Rating is then obtained by taking the arithmetic mean of the site ratings in each iteration. The calculation of the Scenario rating is obtained as above using the average Website rating. The overall WebMark 2004 rating is then obtained from the scenario rating as above.