• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

RAID Question

S

serialb

Diamond Member
Quoted from the RAID article:


<< Typically, it is best to use identical drives. This ensures that both drives react to the RAID environment in the same way. It is possible to use different brands or models on a RAID array, but various issues such as reduced array size and/or decreased performance may result but more on that in the individual sections. >>



What exactly makes the drives &quot;identical&quot;? Assuming both drives are UDMA-66 at 7200 RPM and have identical number of cylinders and such. Here is an example of such a pair of drives: Quantum KA &amp; Quantum KX

They seem &quot;identical&quot; enough to me. But is that enough for a RAID configuration without much loss in terms of performance?

Thanks in advance.

serialb
 
It would seem enough of a difference to me to at least affect the performance of the array.
 
Note that the article said 'typically' not mandatory. Its basically the old weakest link syndrome. The performance of your array is likely to be limited (or driven) by the 'slowest' drive. Consider a striped array with a bunch of very fast drives and 1 dog drive. When you issue a read to that array the command is not completed until all the drives have transferred their data. The dog is likely to be the last one done. The array would perform very similar if the fast drives were all replaced by the same dogs.. If the dog was replaced with a fast drive identical or very similar to the other fast drives then the performance would improve. From a capacity point of view if you have a bunch of big drives and one that is x MB smaller then the software views all the drives as the smallest (to make the striping algorithm work). Standardized capacities in SCSI avoids this problem but Im not sure how well IDE does in this area.
 


<< Note that the article said 'typically' not mandatory. Its basically the old weakest link syndrome. The performance of your array is likely to be limited (or driven) by the 'slowest' drive. Consider a striped array with a bunch of very fast drives and 1 dog drive. When you issue a read to that array the command is not completed until all the drives have transferred their data. The dog is likely to be the last one done. The array would perform very similar if the fast drives were all replaced by the same dogs.. If the dog was replaced with a fast drive identical or very similar to the other fast drives then the performance would improve. From a capacity point of view if you have a bunch of big drives and one that is x MB smaller then the software views all the drives as the smallest (to make the striping algorithm work). Standardized capacities in SCSI avoids this problem but Im not sure how well IDE does in this area. >>



Exactly 🙂
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top