Although I have been using SAS (serial attached SCSI) drives for a few months now, I just recently had a chance to do some side by side comparisons of SAS, SCSI and SATA drives on identical hardware.
The servers were all HP ML350 G4p servers, Xeon 3Ghz processors, 4GB ram, running Server2003 Std. Both the SAS and SATA equipped servers used a HP P600 RAID controller. The SCSI server used a smart array 641. All three were configured with 4 drives in a raid5 configuration with a fifth drive as a hot spare. Both the SCSI and SAS used 72GB drives, the SATA 80GB drives.
As expected the SAS drives out performed both the SCSI and SATA drives. In my simple tests (actually doing baseline performance tests before deployment) the SAS drives were some 40% better than SCSI and over 180% better than SATA.
Have any of you begun using SAS drives? If so what has your experience been? Anybody started using the new 2.5 inch form factor SAS drives?
The servers were all HP ML350 G4p servers, Xeon 3Ghz processors, 4GB ram, running Server2003 Std. Both the SAS and SATA equipped servers used a HP P600 RAID controller. The SCSI server used a smart array 641. All three were configured with 4 drives in a raid5 configuration with a fifth drive as a hot spare. Both the SCSI and SAS used 72GB drives, the SATA 80GB drives.
As expected the SAS drives out performed both the SCSI and SATA drives. In my simple tests (actually doing baseline performance tests before deployment) the SAS drives were some 40% better than SCSI and over 180% better than SATA.
Have any of you begun using SAS drives? If so what has your experience been? Anybody started using the new 2.5 inch form factor SAS drives?