Specifications

A Principled Technologies test report 44
Dell Active System 800 converged infrastructure solution: User
collaboration performance
APPENDIX K – RUNNING THE TEST
To run the test, we configured a 60 minute test window, allowing each benchmark 15 minutes to warm up prior
to the measurement window for performance and latency on each of the solutions.
1. Shut down all VMs, and using the EqualLogic management interface, create a snapshot of each volume. These
snapshots were used to restore to a pre-test state so additional identical runs could be performed.
2. Begin running the LoadGen benchmark. LoadGen was set to run for 1 hour, with a 15-minute allotted warm-up
period, 30-minute performance measurement window, and 15-minute cool-down period.
3. Immediately after beginning the LoadGen benchmark, start the WSSDW workload. This workload was set to run
for 1 hour, with a 15-minute warm up period, 30-minute run period and measurement window, and 15-minute
cool-down period.
4. Immediately after beginning the WSSDW workload, start the three LyncPerfTool workloads using the generated
script. These workloads were set to have a 15 minute warmup time and then run for 30 minutes of recorded
performance.
5. Immediately after beginning all three LyncPerfTool workloads, start the DVD Store benchmark. DVD Store was
set to warm up for 15 minutes, and then run for 30 minutes of recorded performance.
6. On completion of all of the test loads, collect the test results for each test and shut down all of the VMs.
7. Using the EqualLogic management interface, revert to the snapshot of each volume taken before the test
started to restore all databases and workloads to an initial state.
About running the LoadGen test
Microsoft Exchange Load Generator (LoadGen) is an industry-standard tool for benchmarking Microsoft
Exchange Server 2003, 2007, or 2010 mail servers. LoadGen simulates standard mail activity from a set number of users.
When the workload finishes, LoadGen calculates the number of milliseconds necessary to complete its predefined tasks
and reports these response times.
LoadGen measures the impact that mail clients using the MAPI, OWA, IMAP, POP, and SMTP protocols have on
Exchange servers. We can test how effectively a server or guest running Exchange responds to email loads. The
simulator, running on a client machine, generates mail load by sending multiple requests from many users to the
Exchange server.
To run the test, we installed the LoadGen client and server components and simulated 1,000 Exchange users
targeting the Exchange VM using LoadGen’s Average Workload profile. We ran the LoadGen benchmark for 60 minutes,
and beginning at the 15 minute mark, we used a 30 minute measurement window for performance monitoring.
For more details about LoadGen, see
http://www.microsoft.com/download/en/details.aspx?displaylang=en&id=14060.
About running the SharePoint Server test
To simulate the traffic of 1,000 SharePoint Server 2010 users, we used a customized version of the WSSDW
1.0.0.0 Beta test. This test creates sample data, populates the server with it, and simulates SharePoint Server users
completing everyday tasks.