Alan Altmark wrote: >The 3350 is from an era (not so long ago!) when there were no disk "arrays". >No striping. The "location" information in the I/O architecture was physical. >If you took the media out of its housing, you could generally point to where >the data was located. Want protection from drive errors? Write it to another >drive. The good news was that a drive failure was an isolated event. The >drive "next door" wasn't affected.
>Today, the I/O architecture remains, but the mapping of logical to physical >location is an exercise left to the storage unit itself. You can't really >point to where your data is except to point to the storage unit and say, >"Somewhere in there." My experience is that when something dies, you lose >lots of host drives. But physical drive failures have minimal impact due to >striping (RAID). And I'm 99.9% sure that DASD capacity was determined by building the geometry and then trying various densities until error rates became unacceptable, then backing off slightly. Which would explain the weird, random sizes with each generation (until 3390, after which it went to arrays and became standardized-on what future generations will consider a weird size). ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
