gsteckel@diag2.East.Sun.COM (Geoff Steckel - Sun BOS Software) (05/01/90)
In article <PCG.90Apr30221039@odin.cs.aber.ac.uk> pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: >This value of 3600 RPM has been constant for the past 30 years. from >mainframes to micros, and for good (e.g. mechanical) cause. There have >been steady improvements in seek times and bandwidth, and rotational >latency is thus becoming the bottleneck. Palliatives revolve around the >multiple arm idea, i.e. reinventing drums, e.g. disc arrays (introduced >to the 386 world by compaq or zenith recently). In general this is true, but: Some CDC->Imprimis->Seagate drives (including the `Wren Runner', I believe) have gone as high as 5400 RPM. This allows the rotational latency to be reduced to 6 or less milliseconds, at a cost of a 50% higher transfer rate. There may be other manufacturers playing the same game, since this makes a very snappy drive. Also, some of the zone-recording compromises are easier to make at a high data rate rather than a low one. In the other direction, some of the (now obsolete) 9" and 14" SMD drives used 2100 and 2400 RPM to slow down the transfer rate low enough to meet maximum rate limits on old equipment. You could change a belt & pulley arrangement if you wanted faster bits... geoff steckel (gwes@wjh12.harvard.EDU) (...!husc6!wjh12!omnivor!gws) Disclaimer: I am not affiliated with Sun Microsystems, despite the From: line. This posting is entirely the author's responsibility.
gerry@zds-ux.UUCP (Gerry Gleason) (05/02/90)
In article <PCG.90Apr30215541@odin.cs.aber.ac.uk> pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: >In article <511632@nstar.UUCP> larry@nstar.UUCP (Larry Snyder) writes: > I assume that this 6-8 sectors is based on operating systems which > don't have (or are using) a fast file system.. >Not really. The so called "fast file systems" work by *keeping* the free >list sorted by block location (usually with a bitmap), not by block free >time like the V7/SysV one. I think you have missed the real point of FFS. By keeping the free list as a bit array ordered by disk location, you can allocate groups of blocks at a time in contiguous locations. The big win is then in reading sequentially, multiple blocks of a file can be read with a single read operation (of course the same can apply to writes). Due to this behavior, the gap variable becomes almost if not completely meaningless. Gerry Gleason
bill@bilver.UUCP (Bill Vermillion) (05/02/90)
In article <PCG.90Apr30221039@odin.cs.aber.ac.uk< pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: < <In article <1990Apr29.000503.10934@nebulus.UUCP> dennis@nebulus.UUCP <(Dennis S. Breckenridge) writes: < < pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: < < >Essentially the gap size depends on rotational speed of the disc < >(virtually a constant), and on interrupt latency and IO operation < ^^^^^^^^^^^^^^^^^^^^^^ What is this? Different drives have DIFFERENT < speeds. Even the same supplier does NOT adhere to the same speeds. < <Bah. Virtually every drive around does 3600 RPM, for an average <rotational latency of 8.3 ms. Exceptions I know are the optical drives <(slower, they have a limit in the state changes/sec. of their <head/medium). And just last week I was looking over the specs on the current Shugart lines, and there were 3 different drive rotational speeds listed. All were SCSI drives. Since we just talk to the host in SCSI land it really doesn't make any difference to us. And many of the new SCSI's are using zone bit recording - so we don't know much about the data, execpt it's block #. -- Bill Vermillion - UUCP: uunet!tarpit!bilver!bill : bill@bilver.UUCP
root@maxed (0000-Admin(0000)) (05/02/90)
In article <715@bilver.UUCP> bill@bilver.UUCP (Bill Vermillion) writes: >In article <PCG.90Apr30221039@odin.cs.aber.ac.uk< pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: >< ><In article <1990Apr29.000503.10934@nebulus.UUCP> dennis@nebulus.UUCP ><(Dennis S. Breckenridge) writes: >< >< pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: >< >< >Essentially the gap size depends on rotational speed of the disc >< >(virtually a constant), and on interrupt latency and IO operation >< ^^^^^^^^^^^^^^^^^^^^^^ What is this? Different drives have DIFFERENT >< speeds. Even the same supplier does NOT adhere to the same speeds. >< ><Bah. Virtually every drive around does 3600 RPM, for an average ><rotational latency of 8.3 ms. Exceptions I know are the optical drives ><(slower, they have a limit in the state changes/sec. of their ><head/medium). > >And just last week I was looking over the specs on the current Shugart lines, >and there were 3 different drive rotational speeds listed. > >All were SCSI drives. Since we just talk to the host in SCSI land it really >doesn't make any difference to us. And many of the new SCSI's are using zone >bit recording - so we don't know much about the data, execpt it's block #. Many of the new 20 and 25 MHz ESDI & SCSI drives are turning at 5400 rpm. Latency and access times are reduced. Can't count on 3600 any more. -- Ed Whittemore uunet!maxed!ed American Micro Group 201 944 3293
dougp@ico.isc.com (Doug Pintar) (05/02/90)
In article <PCG.90Apr30215541@odin.cs.aber.ac.uk> pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: > >I mean: when the file system is newly built, both the fast file system >and the traditonal one have a free list sorted by location, and give >essentially the same thruput (modulo other possible differences, like >clustering of reads of writes). Just to clear up any confusion, the 386/ix FFS DOES cluster reads, writes, and allocations. Reads of 16KB and more are common. The standard filesystem does separate reads for each block, 1KB at a time. It is the interval between THESE read requests that the FS gap is supposed to cover. >This has *nothing* to do with the filesystem gap; this is a function of >the interval spent by the OS between two disc operations (interrupt >latency, time taken to empty/fill the controller's buffer, time taken to >reissue the IO parameters). If the OS takes one millisecond between two >IO transfers, the disc arm will have flown over typically one sector (of >an MFM drive), and thus an interleave of 1:2 will be advisable. This is actually a case for changing the HARDWARE interleave, not the FS gap. A 1KB block takes up 2 sectors on an AT-type machine, so the mkfs gap won't help with the reading of the 2 (logically) consecutive sectors making up each FS block. >If you have a controller capable of hardware scatter gather, and an OS >that can exploit it, then you don't need to interleave the free list, as >consevutive sectors can be read without any interval for OS processing >between them. The standard AT controller (and clones) is a PIO device! This kind of gives you (admittedly CPU-intensive) scatter/gather that IS used by the HPDD. Any AT-compatible controller that supports 1:1 interleave has buffer memory on board to handle any slack taken up by interrupt processing, data movement, etc. The Adaptec 1542 SCSI adapter supports hardware scatter/gather that is also used by the HPDD. Doug Pintar
sl@van-bc.UUCP (Stuart Lynne) (05/03/90)
In article <715@bilver.UUCP> bill@bilver.UUCP (Bill Vermillion) writes: >In article <PCG.90Apr30221039@odin.cs.aber.ac.uk< pcg@cs.aber.ac.uk (Piercarlo Grandi) writes: >< >< >(virtually a constant), and on interrupt latency and IO operation >< ^^^^^^^^^^^^^^^^^^^^^^ What is this? Different drives have DIFFERENT ><Bah. Virtually every drive around does 3600 RPM, for an average ><rotational latency of 8.3 ms. Exceptions I know are the optical drives >And just last week I was looking over the specs on the current Shugart lines, >and there were 3 different drive rotational speeds listed. >All were SCSI drives. Since we just talk to the host in SCSI land it really Yes, with SCSI the drive manufacturers can finally break loose of the drive speed. With the ST506/ST412 interface MFM/RLL drives the rotation speed is fixed at 3600. When you go to SCSI the interface merely specifies that you can read and write blocks in a range (ie read block m from a range from 0 to n). So the manufacturer can do anything he wants to give you the n blocks. From speeding up the rotation or slowing it down or eliminating it entirely (eg with a RAM disk). It's common on SCSI drives to slow the disk down at the outside of the disk to take advantage of the extra capacity. -- Stuart.Lynne@wimsey.bc.ca ubc-cs!van-bc!sl 604-937-7532(voice) 604-939-4768(fax)