poage@sunny.ucdavis.edu (Tom Poage) (07/19/89)
I have run into an apparently unresolvable question.
Is there a performance hit in having an ESDI disk
hooked to a SCSI bus through a translator (Emulex MD21)
versus a SCSI disk hanging directly on the bus?
ESDI disk performance is supposed to be higher than
that of SCSI disks. Does this mean that the two
configurations are roughly equivalent in performance
after taking into account the MD21?
The questions burn on and on.
Thanks. Tom.
--
Tom Poage, UCDMC Clinical Engineering, Sacto., CA
poage@sunny.ucdavis.edu {...,ucbvax}!ucdavis!sunny!poagehenry@utzoo.uucp (Henry Spencer) (07/20/89)
In article <398@sunny.ucdavis.edu> poage@sunny.ucdavis.edu (Tom Poage) writes: >Is there a performance hit in having an ESDI disk >hooked to a SCSI bus through a translator (Emulex MD21) >versus a SCSI disk hanging directly on the bus? The odds are very good that your "SCSI disk" is in fact an ESDI disk or something similar with such a translator built in. No disk drive directly talks SCSI; there has to be a controller in there somewhere, even if it's built into the drive. Many manufacturers supply the same drive as a SCSI drive or an ESDI (or whatever) drive simply by leaving out the controller board and the SCSI connector if the order doesn't say "SCSI". Any difference in performance will be a matter of controller design, disk data rate, etc., and close attention to the specs is needed. >ESDI disk performance is supposed to be higher than >that of SCSI disks... *If* the ESDI drive is hooked directly to your machine, not via SCSI. Running it through a SCSI translator eliminates any inherent performance gain. (Actually there is no inherent reason why SCSI drives can't be blazing fast, it's just that most of them aren't.) -- $10 million equals 18 PM | Henry Spencer at U of Toronto Zoology (Pentagon-Minutes). -Tom Neff | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
tneale@aeras.UUCP (Tom Neale) (07/20/89)
In article <398@sunny.ucdavis.edu> poage@sunny.ucdavis.edu (Tom Poage) writes: >Is there a performance hit in having an ESDI disk >hooked to a SCSI bus through a translator (Emulex MD21) >versus a SCSI disk hanging directly on the bus? >ESDI disk performance is supposed to be higher than >that of SCSI disks. Does this mean that the two >configurations are roughly equivalent in performance >after taking into account the MD21? Most manufacturers use the same HDA and servo systems on their SCSI and ESDI products of a given form factor and capacity. What this means is the controller is what causes most of the differences in performance. ESDI is a fairly low level interface while SCSI is a higher level, "intelligent" interface. SCSI drives often have lower performance numbers because of the command processing overhead required. This is the time it takes the controller to decode, interpret and execute a command. This time is anywhere from 0.5 to 2.5 milliseconds. Other things that will affect performance of a SCSI drive are transfer rate of the drive *and* controller (sync vs. async), disconnect capability, read ahead caching and so on. The combination of controller and drive will affect performance as well because of the compatibility of SCSI implementations. On a single drive system you probably will see better performance with and ESDI controller and drive. If you have only one controller and multiple drives SCSI stands to give you better overall performance because you can overlap operations. How well will your ESDI drive work? It depends on all the things I've mentioned and probably on a few I've forgotten. How's that for a long winded non-answer? -- Blue skies, | ...sun!aeras!tneale | | in flight: N2103Q | The hurrieder I go Tom Neale | in freefall: D8049 | the behinder I get. | via the ether: WA1YUB |
terryk@pinocchio (Terence Kelleher) (07/20/89)
In article <1989Jul19.170414.20326@utzoo.uucp>, henry@utzoo (Henry Spencer) writes: > >The odds are very good that your "SCSI disk" is in fact an ESDI disk or >something similar with such a translator built in. No disk drive directly >talks SCSI; there has to be a controller in there somewhere, even if it's >built into the drive. Many manufacturers supply the same drive as a SCSI >drive or an ESDI (or whatever) drive simply by leaving out the controller >board and the SCSI connector if the order doesn't say "SCSI". I'll probably get in trouble for this, but not all SCSI drives translate to another standard. The Wren series from CDC/Imprimis/Seagate (whatever they are this week) is clearly a direct SCSI disk. The SCSI drives are formatted using Zone Bit Recording, which modifies the access clock in Zone areas to increase the number of sectors per track on outer cylinders, increasing both capacity and average transfer rate. This is not possible with ESDI, SMD or any other disk specific interface. The SCSI drives have only one controller card. The SCSI and ESDI versions of a given Wren product do not nessecarily have even the same HDA, much less share electronics. > >*If* the ESDI drive is hooked directly to your machine, not via SCSI. >Running it through a SCSI translator eliminates any inherent performance >gain. > >(Actually there is no inherent reason why SCSI drives can't be blazing >fast, it's just that most of them aren't.) If you have multiple disks on your system, the SCSI interface disks may well outperform the ESDI interface disks. The SCSI bus can transfer at a faster rate than the ESDI disks, and since the SCSI controllers typically buffer data, the transfer on the bus does not need to take place in sync with transfer from the heads. Multiple disks can overlap operations and yeild a higher agragate throughput. >-- >$10 million equals 18 PM | Henry Spencer at U of Toronto Zoology >(Pentagon-Minutes). -Tom Neff | uunet!attcan!utzoo!henry henry@zoo.toronto.edu -- Terry Kelleher, Encore Computer Phone: 508-460-0500 UUCP: {bu-cs,decvax,necntc,talcott}!encore!terryk Internet: terryk%pinocchio@multimax.ARPA
jlohmeye@entec.Wichita.NCR.COM (John Lohmeyer) (07/21/89)
In article <398@sunny.ucdavis.edu> poage@sunny.ucdavis.edu (Tom Poage) writes: >I have run into an apparently unresolvable question. >Is there a performance hit in having an ESDI disk >hooked to a SCSI bus through a translator (Emulex MD21) >versus a SCSI disk hanging directly on the bus? > You ask a very difficult question. So much depends on factors that you did not specify. Knowing the system and the controller helps, but not knowing the application probably would render any answer invalid. I say 'probably' because it may be the case that any application would be affected in an adverse way by adding the MD21. But I doubt it. Many SCSI to ESDI controllers actually delivery better performance than an embedded SCSI controller. In fact, it is possible to see better performance than the raw performance specifications of the ESDI drive would seem to indicate. Especially if the MD21 is a caching controller (I am not familiar with it). There is really only one good way to compare performance of two configurations. You have got to benchmark both configurations with a REALISTIC benchmark program. A realistic benchmark is one that accesses the disk in a pattern that is similar to the application you intent to run. In fact, the most realistic benchmark IS the application you intent to run. Although it is often difficult to get objective data from a real application. Unfortunately, our industry has a number of standard benchmark programs that often are poor predictors of actual system performance. I have limited first-hand experience with these, but have heard horror stories. Vendors then often fall into the trap of tuning their systems to run the benchmarks well. The benchmarks then run like a bat-out-of-hell, but the real applications just don't seem to do so well... -- John Lohmeyer J.Lohmeyer@Wichita.NCR.COM NCR Corp. uunet!ncrlnk!ncrwic!entec!jlohmeye 3718 N. Rock Rd. Voice: 316-636-8703 Wichita, KS 67226 SCSI BBS 316-636-8700 300/1200/2400 24 hours