White Papers

12 BP1062 | Best Practices for Decision Support Systems with Microsoft SQL Server 2012 using Dell EqualLogic PS Series
Storage Arrays
4.2 High-level system design
A high-level overview of the test configuration is shown in Figure 2.
High-level overview of the test configuration Figure 2
Key design details of the test system configuration shown in Figure 2 include:
• Two R820 Dell PowerEdge servers were used to host two SQL Server instances. Each of these
servers had SQL Server 2012 Enterprise Edition installed on Windows Server 2012 Enterprise
Edition.
• Both of the servers had 128 GB RAM each, and the SQL server instances were configured to use
117.7 GB of the memory by using the maximum server memory setting in SQL Server
Management Studio.
• The infra server illustrated in Figure 2 had VMware ESXi 5 installed on it and hosted virtual
machines for vCenter and active directory.
• The load gen server in Figure 2 had VMware ESXi 5 installed on it and hosted four Windows 2008
R2 workload simulation virtual machines with each running an instance of Benchmark Factory.
• The monitor server was a PowerEdge R710 running Windows 2008 R2 Enterprise edition. It
hosted SAN HQ.
• The SAN switches consisted of two Dell Force10 S4810 switches configured with a LAG
(2 x 40 GbE) interconnect. Redundant connection paths were created from each array controller
to each switch in the stack.
• Two EqualLogic PS6110X arrays, each with 24 x 900 GB 10K RPM SAS disk drives in a RAID 50
configuration hosted SQL Server database volumes. For single array tests (sections 5 to 6.3
), the
two members were placed in separate pools and for the two arrays scalability tests (sections
6.4.3 and 6.4.4) the two members were placed in the same pool.