Specifications

A Principled Technologies test report 2
Hardware upgrades to improve database, SharePoint, Exchange, and
file server performance with the Intel processor-powered Dell
PowerEdge T630
BETTER PERFORMANCE WITH THE DELL POWEREDGE T630
Businesses running tower servers purchased four or more years ago must work
with aging hardware that can limit resources and constrict user growth. New servers,
such as the Dell PowerEdge T630 tower server, powered by Intel Xeon E5-2660 v3
processors, offer robust hardware to help avoid application performance deterioration
and have designs that promote component upgrades.
We tested the PowerEdge T630 several ways. First, we compared a base
configuration of the server to a four-year-old legacy server. We set up the following
solutions:
 An Intel Xeon processor E5-2660 v3-based Dell PowerEdge T630 as our
base configuration, running Windows Server® 2012 R2 with Hyper-V®
 An HP ProLiant ML350 G6 as our legacy server, running Windows Server
2008 R2 with Hyper-V
To demonstrate the performance benefits of migrating workloads to the new
Dell tower server, both solutions had virtual instances of Microsoft® SQL Server® 2012,
Microsoft Exchange Server 2010, SharePoint Server 2010, and a file server, each on its
own Windows Server 2008 R2 VM. In this comparison, we simulated the workload of
200 users each for Exchange, SharePoint, and file server and ran a heavy SQL database
workload. Our Exchange and SharePoint VMs each had to support their 200 users with
an acceptable latency (under 100ms for Exchange and under one second for
SharePoint). Our file server VM needed to perform a minimum of 0.3 IOPS per user, or
60 IOPS for 200 users. Each of our database VMs needed to deliver 14,000 orders per
minute (OPM) on average. We measured Exchange performance using Exchange Load
Generator 2010, SharePoint performance using WSSDW 1.0, database performance with
DVD Store 2.1, and file server performance using Iometer. For more information on
what and how we tested, see Appendix A.
In our second set of tests, we compared the base configuration of the
PowerEdge T630 against configurations with upgraded hardware components to see
how workload performance increased at every step. Each upgraded configuration saw a
new component in addition to the component from the previous step, introducing a
new way to increase performance in a mixed workload environment. We upgraded with
the following components:
 Memory – upgrade to 384 GB of RAM
 Drive type – upgrade to four 800GB SSDs
For detailed information on the test systems, see Appendix B; for detailed steps
on how we tested, see Appendix C; and for detailed results, see Appendix D.