Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

SAS drives can address multiple IO operations from a single platter revolution if the data happens to be where the head is already positioned. How can you measure whether this actually makes a difference, comparing SATA vs. SAS, especially with a good raid card?

I have a Dell PE with a Perc 5, eight drive SATA raid. I can either replace SATA with SAS and the same raid card or replace the raid card and purchase new SATA drives. I have other servers that have SAS and wanted to know how I could measure (or if software exists to measure) the actual benefit of SAS over SATA?

share|improve this question
2  
Could you be much more specific as to how you're measuring performance and what kind of loads you're subjecting the array. Each type has it's own performance focus: SATA->Space, SAS->IO Throughput/Longevity balance, SSD->IO Throughput. –  Chris S May 16 '11 at 4:24
    
Some SATA drives support Native Command Queuing (the feature you're referencing), too. –  Shane Madden May 16 '11 at 14:07

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.