Readers will find a BigAdmin report of the test conducted on the appropriateness of the "Sun StorageTek 6540 Array Storage Solution for 20,000 Mailboxes in Microsoft Exchange 2007," as the study is entitled. It provides information on a Sun Microsystems storage solution for Microsoft Exchange Server, based on the Microsoft Exchange Solution Reviewed Program (ESRP).

This Microsoft Exchange 2007 primary storage solution was tested on the Sun StorageTek 6540 array with the targeted number of users set at 20,000 with four servers. The tested user profile was 0.5 (0.42 + 20%) Input/Output Per Second (IOPS) per mailbox with a limit of 300 Mbytes.

The tested solution consists of a Sun StorageTek 6540 dual Fibre Channel (FC) RAID controller cabinet with 10 Sun StorageTek CSM200 rack-ready expansion tray modules. There are one hundred sixty 146-Gbyte 15,000-rpm, FC 4-Gbps drives for 23.36 Tbytes (raw) / 11.68 Tbytes (usable, RAID 10). This solution can expand to 14 StorageTek CSM200 modules for a total of 168 Tbytes (raw) / 84 Tbytes (usable, RAID 10) with 750-Gbyte SATA-II drives.

Key specifications include:

• Dual Fibre Channel RAID controller
• End-to-end speed of 4 Gbps
• Ability to scale up to 168 Tbytes and 224 drives
• 575,000 IOPS out of cache
• Supports high-performance FC and high-capacity SATA drives

A number of tests in the framework check reliability, and these tests run for 24 hours, the goal being to verify the storage can handle high I/O load for a long period of time. Both log and database files were analyzed for integrity after the stress test to ensure no database or log corruption. No errors were reported in the saved eventlog file, nor were any errors reported during the database and log checksum process.

Primary Storage Performance Results

The primary storage performance testing is designed to exercise the storage with maximum sustainable Microsoft Exchange type of I/O for 2 hours. The test shows how long it takes for the storage to respond to an I/O under load. The data shown below show the sum of all the logical disk I/O operations and an average of all the logical disks I/O latency in the 2-hour test duration. Each server is listed separately and the aggregate numbers across all servers are listed as well.

Data in the next paragraph show the individual server metrics: the sum of I/O operations across storage groups and the average latency across all storage groups on a per-server basis.

Database I/O

Average database disk transfers/sec 3202.92 Average database disk reads/sec 1773.321 Average database disk writes/sec 1429.599 Average database disk read latency 12 ms Average database disk write latency 7 ms

Transaction Log I/O

Average log disk writes/sec 830.962 Average log disk write latency 1 ms

Database I/O

Average database disk transfers/sec 3429.772 Average database disk reads/sec 1892.125 Average database disk writes/sec 1537.647 Average database disk read latency 11 ms Average database disk write latency 8 ms

Transaction Log I/O

Average log disk writes/sec 882.556 Average log disk write latency 1 ms

Database I/O

Average database disk transfers/sec 2961.134 Average database disk reads/sec 1633.004 Average database disk writes/sec 1328.130 Average database disk read latency 11 ms Average database disk write latency 8 ms

Transaction Log I/O

Average log disk writes/sec 768.327 Average log disk write latency 1 ms

Database I/O

Average database disk transfers/sec 3228.218 Average database disk reads/sec 1779.871 Average database disk writes/sec 1448.346 Average database disk read latency 12 ms Average database disk write latency 6 ms

Transaction Log I/O

Average log disk writes/sec 839.876 Average log disk write latency 1 ms

The following data show the sum of I/O operations across all servers in the solution and the average latency across all servers in the solution.

Database I/O

Average database disk transfers/sec 12822.044 Average database disk reads/sec 7078.321 Average database disk writes/sec 5743.722 Average database disk read latency 11.5 ms Average database disk write latency 7.25 ms Transaction Log I/O Average log disk writes/sec 3321.721 Average log disk write latency 1 ms

For the Version 1.0 release, only streaming backup type is supported for testing in the framework. There are two tests in this section. The first one measures the read I/O performance metrics by running a checksum on all the databases and log files. The second test measures the end-to-end performance when the databases are backed up to disks.

Database Read-only Performance

The test measures the maximum rate at which databases could be streaming backed up with the following data showing the average rate for a single database file:

I/O Read Performance

MB read/sec per storage group 46.36 MB read/sec per system 324.538 MB read/sec total 1298.15

Log Read-only Performance

This test measures the maximum rate at which the log files can be played against the databases. Data show the average rate for 500 log files played in a single storage group. Each log file is 1 Mbyte.

End-to-End Performance

Average time to play one log file (sec) .362

As in the case of the study involving the Sun StorageTek 6140 Array [20659], the BigAdmin paper cautions that "the report was developed by storage solution providers and reviewed by the Microsoft Exchange Product team. The test results and data presented in this document are based on the tests introduced in the ESRP test framework. Customers should not quote the data directly for their pre-deployment verification. It is still necessary to go through the exercises to validate the storage design for a specific customer environment."

		fullsource
Print Email Current Profile Archive Blog Generate newsletter