carpent@coltrane.SRC.Honeywell.COM (Todd Carpenter) (04/30/89)
In case you folks are interested, I ran DiskPerf on my beastie:
B2000 Rev 4.4
A2090a
A2620
Rodime SCSI HD. Basically, it is their 180MB 5 1/4" SCSI drive, with a few
hand modifications. Like, 12msec access, 12MHz transfer. Hee-hee.
Results:
File create/delete: create 18 files/sec, delete 52 files/sec
Directory scan: 156 entries/sec
Seek/read test: 135 seek/reads per second
r/w speed: buf 512 bytes, rd 80556 byte/sec, wr 29289 byte/sec
r/w speed: buf 4096 bytes, rd 197844 byte/sec, wr 171897 byte/sec
r/w speed: buf 8192 bytes, rd 327680 byte/sec, wr 257846 byte/sec
r/w speed: buf 32768 bytes, rd 635500 byte/sec, wr 372275 byte/sec
r/w speed: buf 131072 bytes, rd 655360 byte/sec, wr 427990 byte/sec
r/w speed: buf 524288 bytes, rd 699050 byte/sec, wr 455902 byte/sec
Unfortunately, I was surprised the numbers were so low. I had hoped for
significantly higher. Anyone have any suggestions? Or are these fast for Amiga
drives?elg@killer.Dallas.TX.US (Eric Green) (05/01/89)
In message <21149@srcsip.UUCP>, carpent@coltrane.SRC.Honeywell.COM (Todd Carpenter) says: >In case you folks are interested, I ran DiskPerf on my beastie: >A2090a >A2620 >Rodime SCSI HD. Basically, it is their 180MB 5 1/4" SCSI drive, with a few > hand modifications. Like, 12msec access, 12MHz transfer. Hee-hee. >r/w speed: buf 32768 bytes, rd 635500 byte/sec, wr 372275 byte/sec >r/w speed: buf 131072 bytes, rd 655360 byte/sec, wr 427990 byte/sec >r/w speed: buf 524288 bytes, rd 699050 byte/sec, wr 455902 byte/sec >Unfortunately, I was surprised the numbers were so low. I had hoped for >significantly higher. Anyone have any suggestions? Or are these fast for Amiga >drives? I suspect that the limiting factors are a) the file system, and b) the A2090A. I got about the same results from a 2090A and a Quantum 80S. The 2090A limits speed because it is limited to the old 1mb/sec asynchronous SCSI protocol (meaning that however hot your drive is, it's limited to 1mb/sec max throughput). However, someone else got basically the same numbers from a Quantum 80s and a Microbotics Hardframe (which has the new 5mb/sec synchronous SCSI, i.e., will take data as fast as the Quantum can dish it out, 4mb/sec). So I have to conclude that file system overhead has something to do with it, too.... -- | // Eric Lee Green P.O. Box 92191, Lafayette, LA 70509 | | // ..!{ames,decwrl,mit-eddie,osu-cis}!killer!elg (318)989-9849 | | \X/ Newsflash: DP director fired for buying IBM! |
steveb@cbmvax.UUCP (Steve Beats) (05/01/89)
In article <7965@killer.Dallas.TX.US> writes: > >I suspect that the limiting factors are a) the file system, and b) the >A2090A. I got about the same results from a 2090A and a Quantum 80S. >The 2090A limits speed because it is limited to the old 1mb/sec >asynchronous SCSI protocol (meaning that however hot your drive is, >it's limited to 1mb/sec max throughput). However, someone else got >basically the same numbers from a Quantum 80s and a Microbotics >Hardframe (which has the new 5mb/sec synchronous SCSI, i.e., will take >data as fast as the Quantum can dish it out, 4mb/sec). So I have to >conclude that file system overhead has something to do with it, >too.... > >-- While it's true that the filing system will eat some of the performance from a SCSI drive, it's not reasonable to expect a synchronous mode controller to run at 5.0Mb/s. Synchronous mode is relatively new (to controller vendors, not the SCSI spec) and as such it can become quite a pain to support. I attempted to bring up a driver with full sync xfer support and found that a LOT of drives simply crashed when given a sync xfer request. For this reason I had to stop initiating the message and just handle it when the target sent the request first, (an extremely rare occurence). Even if you get a drive that can support full speed synchronous transfers, you are NOT going to get 5.0Mb/s transfer rates. Most drives can pull data off the media at 1.25Mb/s at 1:1 interleave. That is the maximum throughput you are going to get. The only time that you can up this figure is if you have multiple units on the controller, all running synchronous mode and all supporting full disconnect reselect protocols. In this scenario you do want to burst data out of the drives internal cache buffer as fast as you can. While you're bursting out one batch of data, another drive can be filling it's buffer ready for when it reselects the initiator. Even given a bunch of targets (say 4) all doing full speed sync transfers, you will still be limited to the Amigas bus speed. With a 16 bit DMA engine on a stock a2000 this will come out to around 3.58Mb/s. That's the ABSOLUTE maximum data throughput you are going to get no matter how fast the drive can supply data. The fastest drive I have found to date is the CDC Wren V That baby can sustain 1400K/s direct from the drive. Steve
doug@xdos.UUCP (Doug Merritt) (05/02/89)
In article <6720@cbmvax.UUCP> steveb@cbmvax.UUCP (Steve Beats) writes: >will still be limited to the Amigas bus speed. With a 16 bit DMA engine on >a stock a2000 this will come out to around 3.58Mb/s. That's the ABSOLUTE >maximum data throughput you are going to get no matter how fast the drive >can supply data. The fastest drive I have found to date is the CDC Wren V >That baby can sustain 1400K/s direct from the drive. Let me get this straight: Are you saying that the Wren V can sustain 1400K/sec on an *Amiga* or on some other system? If you meant on the Amiga, then I'd like to hear more details, because that's a better number than anything else I've heard. Doug -- Doug Merritt Member, Crusaders for a Better Tomorrow {pyramid,apple}!xdos!doug Professional Wildeyed Visionary "Of course, I'm no rocket scientist" -- Randell Jesup, Capt. Boinger Corps
thad@cup.portal.com (Thad P Floryan) (05/02/89)
Gee, Todd Carpenter thinks his drive is slow, with (using the 8192 byte buffer): read @ 327,680 bytes/second, and write @ 257,846 bytes/second ?? Sheesh! Peruse the April 1989 UNIX Review (Vol. 7, No. 4), page 98, for disk perfs of some "other" systems (these are sequential read tests of the first 2MB of an 8MB file using an 8192 byte buffer): Apple Mac IIx 109K bytes/sec (SCSI) Compaq 386/25MHz 154K bytes/sec Sun 3/50 232K bytes/sec (SCSI) Sun 3/160 237K bytes/sec (SCSI) IBM RT 267K bytes/sec Sun 3/260 620K bytes/sec MIPS M/120-5 656K bytes/sec (SCSI) Sun 4/260 794K bytes/sec Thad Floryan [ thad@cup.portal.com (OR) ..!sun!portal!cup.portal.com!thad ]
elg@killer.Dallas.TX.US (Eric Green) (05/03/89)
in article <6720@cbmvax.UUCP>, steveb@cbmvax.UUCP (Steve Beats) says: $ In article <7965@killer.Dallas.TX.US> writes: $>I suspect that the limiting factors are a) the file system, and b) the $>A2090A. I got about the same results from a 2090A and a Quantum 80S. $>The 2090A limits speed because it is limited to the old 1mb/sec $>asynchronous SCSI protocol (meaning that however hot your drive is, $ While it's true that the filing system will eat some of the performance from $ a SCSI drive, it's not reasonable to expect a synchronous mode controller to $ run at 5.0Mb/s. Synchronous mode is relatively new (to controller vendors, Too true. But going synchronous can probably get you to the magic 1mb/sec Diskperf ;-). (actually, anything over 500k/sec is probably just window dressing as far as most Amiga users will be able to tell, but you must admit that it DOES look neat on advertising ;-). $ not the SCSI spec) and as such it can become quite a pain to support. I $ attempted to bring up a driver with full sync xfer support and found that a $ LOT of drives simply crashed when given a sync xfer request. For this reason $ I had to stop initiating the message and just handle it when the target sent $ the request first, (an extremely rare occurence). Was this driver, by any chance, for the 2090A? Just curious... the 2090 is a fast controller, under good conditions (i.e. DMA'ing into FAST), but it IS a rather old design. Didn't know they had sync SCSI back then. Doesn't surprise me to hear that a lot of drives crashed given the sync xfer request. From what I've heard on the net, most drives barely support standard SCSI, much less extensions. That's what comes of letting the engineers write the code ;-). $ are NOT going to get 5.0Mb/s transfer rates. Most drives can pull data off $ the media at 1.25Mb/s at 1:1 interleave. That is the maximum Note that some drives, e.g. the Quantum (?), have a track buffer. What this means is that if you request block 1025 and then two sectors later find that you need block 1020, it's already there, and can be sent "ASAP" so to speak. Of course, you had to wait for the disk to rotate all the way to 1025 in the first place.... $ will still be limited to the Amigas bus speed. With a 16 bit DMA engine on $ a stock a2000 this will come out to around 3.58Mb/s. That's the ABSOLUTE $ maximum data throughput you are going to get no matter how fast the drive $ can supply data. The fastest drive I have found to date is the CDC Wren V $ That baby can sustain 1400K/s direct from the drive. I think 3.58MB/s is plenty fast ;-). (note what I said about all speeds > 500K/sec... massive overkill!). -- | // Eric Lee Green P.O. Box 92191, Lafayette, LA 70509 | | // ..!{ames,decwrl,mit-eddie,osu-cis}!killer!elg (318)989-9849 | | // Join the Church of HAL, in worship of all computers with | |\X/ three-character names (e.g. IBM and DEC). White lab coats optional. |
lphillips@lpami.wimsey.bc.ca (Larry Phillips) (05/03/89)
In <7965@killer.Dallas.TX.US>, elg@killer.Dallas.TX.US (Eric Green) writes: >In message <21149@srcsip.UUCP>, carpent@coltrane.SRC.Honeywell.COM (Todd Carpenter) says: >>In case you folks are interested, I ran DiskPerf on my beastie: >>A2090a >>A2620 >>Rodime SCSI HD. Basically, it is their 180MB 5 1/4" SCSI drive, with a few >> hand modifications. Like, 12msec access, 12MHz transfer. Hee-hee. >>r/w speed: buf 32768 bytes, rd 635500 byte/sec, wr 372275 byte/sec >>r/w speed: buf 131072 bytes, rd 655360 byte/sec, wr 427990 byte/sec >>r/w speed: buf 524288 bytes, rd 699050 byte/sec, wr 455902 byte/sec >>Unfortunately, I was surprised the numbers were so low. I had hoped for >>significantly higher. Anyone have any suggestions? Or are these fast for Amiga >>drives? >I suspect that the limiting factors are a) the file system, and b) the >A2090A. I got about the same results from a 2090A and a Quantum 80S. >The 2090A limits speed because it is limited to the old 1mb/sec >asynchronous SCSI protocol (meaning that however hot your drive is, Async SCSI specs 1.25 Mbytes/second >it's limited to 1mb/sec max throughput). However, someone else got Andy Finkel tells me he got 1.2 MBytes/second from a CDC Wren V (or possibly a VI), sometime not long after I told him about the 875K/sec from a Wren IV (empty 50 meg partition). >basically the same numbers from a Quantum 80s and a Microbotics >Hardframe (which has the new 5mb/sec synchronous SCSI, i.e., will take Well, they will have it anyway. I called them last week, and it does not yet support synchronous SCSI >data as fast as the Quantum can dish it out, 4mb/sec). So I have to >conclude that file system overhead has something to do with it, >too.... I'm sure it does. The 875K/sec dropped off to about 690K/sec after purring some files on the drive. Diskperf is also quite sensitive to what else is running, and to exactly what the bitmap looks like when you run it. -larry -- Frisbeetarianism: The belief that when you die, your soul goes up on the roof and gets stuck. +----------------------------------------------------------------------+ | // Larry Phillips | | \X/ lphillips@lpami.wimsey.bc.ca or uunet!van-bc!lpami!lphillips | | COMPUSERVE: 76703,4322 | +----------------------------------------------------------------------+
doug@xdos.UUCP (Doug Merritt) (05/07/89)
In article <7985@killer.Dallas.TX.US> elg@killer.Dallas.TX.US (Eric Green) writes: >I think 3.58MB/s is plenty fast ;-). (note what I said about all >speeds > 500K/sec... massive overkill!). I don't see this. Why do you say > 500K/sec is overkill??? My experience over the years with disks on Unix says that faster is better. Granted the access time is usually the dominating factor, but not always. What about the case (either UNIX *or* AmigaDos) when you've carefully arranged to have large files arranged more-or-less contiguously? Then the transfer rate makes a difference. What if you want to dump animation/sound off disk, and you have a huge amount of data. A factor of two in disk transfer could translate directly into a factor of two in the frame rate. The difference in non-animation may be less dramatic but still noticeable. If I load a 1 megabyte executable, or image (etc) then the difference between .5 sec and 1 second will often be noticeable. It won't always be swamped by the access time. Doug -- Doug Merritt Member, Crusaders for a Better Tomorrow {pyramid,apple}!xdos!doug Professional Wildeyed Visionary "Of course, I'm no rocket scientist" -- Randell Jesup, Capt. Boinger Corps
rcj2@cbnewsd.ATT.COM (ray.c.jender) (09/01/89)
Once upon a time there was someone looking for diskperf results
for a Hardframe 2000/Seagate ST157N configuration. Well, here's some:
Hardware: A2000, Microbotics 8-Up/2Meg, Microbotics Hardframe,
Seagate ST157N (48Meg, 3.5").
Software: VirusX, Qmouse.
File create/delete: create 16 files/sec, delete 45 files/sec
Directory scan: 102 entries/sec
Seek/read test: 110 seek/reads per second
r/w speed: buf 512 bytes, rd 76725 byte/sec, wr 27967 byte/sec
r/w speed: buf 4096 bytes, rd 174278 byte/sec, wr 132451 byte/sec
r/w speed: buf 8192 bytes, rd 272357 byte/sec, wr 171897 byte/sec
r/w speed: buf 32768 bytes, rd 400729 byte/sec, wr 206955 byte/sec
r/w speed: buf 131072 bytes, rd 495390 byte/sec, wr 295373 byte/sec
r/w speed: buf 524288 bytes, rd 499321 byte/sec, wr 293993 byte/sec
Hardware: Same as above.
Software: VirusX, Qmouse, The Basic Demo from WB1.3 Extras, (the one
with the four windows of lines, polygons, circles
and animation)
File create/delete: create 10 files/sec, delete 32 files/sec
Directory scan: 15 entries/sec
Seek/read test: 25 seek/reads per second
r/w speed: buf 512 bytes, rd 18493 byte/sec, wr 19007 byte/sec
r/w speed: buf 4096 bytes, rd 115864 byte/sec, wr 88239 byte/sec
r/w speed: buf 8192 bytes, rd 180788 byte/sec, wr 116941 byte/sec
r/w speed: buf 32768 bytes, rd 316153 byte/sec, wr 173318 byte/sec
r/w speed: buf 131072 bytes, rd 436906 byte/sec, wr 271183 byte/sec
r/w speed: buf 524288 bytes, rd 473041 byte/sec, wr 285975 byte/sec
I don't know what version of diskperf i have.
So, all you hardware jocks out there, is this good, bad or ugly?