Specifications

11
Principled Technologies, Inc.: Virtualized Exchange workload performance comparison of end-to-end
solutions with iSCSI storage connections: Dell PowerEdge 2950 with Dell EqualLogic PS5000 vs. HP
ProLiant DL385 with HP Stora
g
eWorks EV
A
4400
15. Once installation is complete, click Finish.
16. Download and install Load Generator using all defaults.
Preparing Load Generator
1. Log into the MailClient using the tester account.
2. Select StartÆAll ProgramsÆMicrosoft ExchangeÆExchange Load Generator.
3. When the Load Generator window appears, select Start a new test.
4. Select Create a new test configuration, and click Continue.
5. Enter tester as the account name.
6. In the Specify test settings window, type Password1 as the Directory Access Password and Mailbox
Account Master Password, and click Continue with recipient management.
7. Make 500 users in each Mailbox Database, and click Continue.
8. To accept defaults for Advanced recipient settings, click Continue.
9. In the Specify test user groups window, select the plus sign to add a user group.
10. Change the Client Type to Outlook 2007 Online and the Action Profile to Custom, and click Continue.
11. Leave defaults in Remote configurations, and click Continue.
12. Click Save the configuration file as, and name it Loadgencfg.
13. Click Start the initialization phase (recommended before running the test) to start initialization. The
initialization process might take a few hours.
Power measurement procedure
To record each server’s power consumption during each test, we used an Extech Instruments (www.extech.com)
380803 Power Analyzer/Datalogger. We connected the power cord from the server under test to the Power
Analyzer’s output load power outlet. We then plugged the power cord from the Power Analyzer’s input voltage
connection into a power outlet. For the storage arrays, which have multiple power cords, we plugged all of the
cords for each array into a power strip and then plugged the power strip into the Power Analyzer’s output load
power outlet.
We used the Power Analyzer’s Data Acquisition Software (version 2.11) to capture all recordings. We installed the
software on a separate Intel processor-based PC, which we connected to the Power Analyzer via an RS-232
cable. We captured power consumption at one-second intervals.
To gauge the idle power usage, we recorded the power usage for 2 minutes while each system was running ESX
Server 3i but otherwise idle, meaning no VMs were powered on.
We then recorded the power usage (in watts) for each system during the testing at 1-second intervals. To
compute the average power usage, we averaged the power usage during the time the system was producing its
peak performance results. Since we measured the power of the server and storage separately, we added the
average power of both together to get the total power. We call this time the power measurement interval. See
Figure 5 (idle and average peak power) for the results of these measurements.
Testing procedure
To perform the test, we used a series of batch files. First, we powered on all virtual machines. We made each wait
for the controller to copy a file to begin the test. Before beginning the test, we let all equipment sit for 1 hour and
30 minutes to ensure the virtual machines were idle and had finished running any background tasks. Then we
executed the start batch file on our controller, which copied a file to all virtual machines to start LoadGen. After
LoadGen finished, we used another batch file to copy the results from all virtual machines. The batch files simply
used the copy command and referred to all virtual machines by their IP addresses. Once a run finished, we used
a restore batch file, which restored the mailstore to the original state. This allowed us to always start from the
same point in testing.