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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Jay who wrote (54681)	4/28/1998 11:46:00 AM
From: rudedog	Respond to of 186894

Fast/wide ultraSCSI has twice the sustained bandwidth of F/W. This is usually not the bottleneck. The advantage to a combination of faster PCI bus and faster SCSI is that small blocks of data (like the NT page faults I was talking about before) may transfer in a single context switch. The time taken by the processor for context switching can be greater than actual work for these small block transfers. The biggest performance bottleneck on disk access is latency, not transfer rate. As Michael said, a RAID card is a huge help with latency. I use a CPQ Smart-2 RAID controller with 4-disk striping and see average latency at the application level of about 3 MSec, even though the disks are about 9MSec latency. Also get transparent fault tolerance, I can take a disk out of the system with no loss of data (but a little performance hit). Hardware RAID is much faster than software RAID and also improves cache performance.