Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : How high will Microsoft fly? -- Ignore unavailable to you. Want to Upgrade?

To: i-node who wrote (62395)	11/1/2001 8:29:13 AM
From: rudedog	Respond to of 74651

David - the point of RAID 0 (hardware striping) is to use more than one drive at a time in a given request. The large transfer rig I use draws from four drives at once, so the transfer rates are 4X that of a single drive. While the drive is indeed the bottleneck, it is much less so in this setup since each drive has its own 100MB channel (as opposed to SCSI where all drives share the pipe). Using very large 10K drives also improves transfer as there is higher density per track. Obviously there are lots of other tuning optimizations needed to maintain that bandwidth, but I can move a 100MB file in about 2 seconds. A big SCSI array is still the fastest way to move small blocks in a multi-tasking environment (for example database access). For that I use a dual channel array controller with 6 15K SCSI drives in RAID 1+0 configuration. That gives me seek times which are can be as low as 1/10 of the individual drive seek times, since the controller will do nearest head seek and if the request queue is deep enough, there is usually something to go after across the whole array. An additional benefit is that the data is all redundantly stored, so a single drive failure does not cause a loss of data. The controller and array configuration has a lot more to do with performance than the individual drive seek times or transfer rates.