| System Hardware | |
|---|---|
| Hardware | Details |
| CPU | Intel Core i7 920 (45 nm, 2.66 GHz, 8 MB L3 Cache) |
| Motherboard | Supermicro X8SAX (Socket LGA 1366), Rev. 1.0 Chipset: Intel X58 + ICH10R BIOS: 1.0B |
| RAM | 3 x 1 GB DDR3-1333 Corsair CM3X1024-1333C9DHX |
| System Hard Drive | Seagate NL35, 400 GB ST3400832NS 7,200 RPM, SATA/150, 8 MB Cache |
| Power Supply | OCZ EliteXstream 800W OCZ800EXS-EU |
| Benchmarks | |
| Performance measurements | h2benchw 3.12 |
| I/O Performance | IOMeter 2006.07.27 Database -Benchmark Fileserver-Benchmark Webserver-Benchmark Workserver-Benchmark |
| System Software and Drivers | |
| Driver | Details |
| Operating System | Windows Vista Ultimate SP1 |
| Intel Chipset Drivers | 9.1.0.1007 |
| AMD Graphics Drivers | Radeon 8.12 |
| Intel Matrix Storage Drivers | 8.7.0.1007 |
Test Procedure
In an effort to really stress the flash SSDs, we ran the following cycle on all drives:
1st cycle
- h2benchw sequential read/write test
- IOMeter fileserver, database, webserver, workstation profile I/O test
2nd cycle
- h2benchw sequential read/write test
- IOMeter fileserver, database, webserver, workstation profile I/O test
3rd cycle
- h2benchw sequential read/write test
- IOMeter fileserver, database, webserver, workstation profile I/O test
4th cycle
- h2benchw sequential read/write test
5th cycle
- h2benchw sequential read/write test
6th cycle
- h2benchw sequential read/write test
We decided to run the sequential h2benchw test and immediately switch to I/O loads three times in a row, so the drives would have to switch from maximum throughput to maximum I/O performance repeatedly. Then we ran three additional h2benchw tests to see if and when the flash SSDs returned to their initial performance levels.
Sponsored
