Specifications

A Principled Technologies test report 28
The datacenter in a box performance test: Comparing Dell and HP
blade solutions
 WSSPutDocumentCoded 10%
r. Click OK.
s. At the Network types screen, select LAN, and click Next.
t. At the Browser screen, click Next.
u. At the Specify computers to monitor with counters sets during load test run screen, click Next.
v. At the Review and edit run settings for load test screen, enter 15 minutes for the Warm-up duration, enter
30 minutes for Run duration, 15 minutes for the Cool down duration, and click Finish.
6. To ensure that the WSSDW workload evenly selecta different SharePoint sites at random, create .csv lists of a
number of categories within each test and addthe following parameters to each:
 WSSD WSSDispForm: Site name, list name, and list IDs.
 WSSDocHTTPFetch: Site name, doc revision, document library name, and document name.
 WSSHomePage: Site name
 WSSSmallDocLibAllItems: Site name, and document library name
 WSSPutDocumentCoded: Site name
7. Add the following lines of code to the WSSPutDocumentCoded test:
[DeploymentItem("officeservertransactions\\siteNames.csv",
"officeservertransactions")]
[DataSource("Sites", "Microsoft.VisualStudio.TestTools.DataSource.CSV",
"|DataDirectory|\\officeservertransactions\\siteNames.csv",
Microsoft.VisualStudio.TestTools.WebTesting.DataBindingAccessMethod.Random,
Microsoft.VisualStudio.TestTools.WebTesting.DataBindingSelectColumns.SelectO
nlyBoundColumns, "siteNames#csv")]
[DataBinding("Sites", "siteNames#csv", "siteName",
"Sites.siteNames#csv.siteName")]
At the "WebTestRequest request1" line, add:
+ this.Context["Sites.siteNames#csv.siteName"].ToString() +
Running the test
To run the test, we staggered the workload, allowing each benchmark time to warm up prior to measuring the
performance and latency on each of the solutions.
1. Begin running the LoadGen benchmark. LoadGen is set to run for 1 hour and 15 minutes, with a 15-minute
allotted warm-up period, 30-minute time of recorded latency, and 15-minute cool-down period.
2. Immediately after beginning the LoadGen benchmark, start the WSSDW workload. This workload is set to run for
1 hour and 15 minutes, with a 15-minute warm up period, 30-minute run period, and 15-minute cool-down
period.
3. Five minutes after LoadGen and WSSDW begin, start the DVD Store benchmark. DVD Store is set to warm up for
10 minutes, and then run for 30 minutes of recorded performance to coincide with LoadGen.