Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!rutgers!sri-spam!ames!ucbcad!ucbvax!ARIZMIS.BITNET!JMS From: JMS@ARIZMIS.BITNET (You do their work, and they shall have good ...) Newsgroups: comp.os.vms Subject: RE: 6250 tape backup Message-ID: <8706031526.AA24352@ucbvax.Berkeley.EDU> Date: Tue, 2-Jun-87 20:09:00 EDT Article-I.D.: ucbvax.8706031526.AA24352 Posted: Tue Jun 2 20:09:00 1987 Date-Received: Sun, 7-Jun-87 01:08:58 EDT Sender: daemon@ucbvax.BERKELEY.EDU Distribution: world Organization: The ARPA Internet Lines: 51 DNEIMAN%CARLETON.EDU@RELAY.CS.NET asks: >Is is possible from the above information to determine whether and how much >the 6250 tape drive will speed up backup (Is the bottle neck tape throughput, >disk i/o, or cpu?). >More info: This drive may or may not be placed on the HSC; it may go on >one of the other machines. Is a 750 a reasonable place for a 6250 bpi drive? >How about a 780? What are the performance tradeoffs? At the University of Arizona we have just converted from dual TU77s to dual TU78s. The driving CPU is a 780; the disks are RA81s on an HSC50. Our observation is that the 780 CPU is a bottleneck for using BACKUP in the canonical fashion (with ECC). We have not seen any real significant speed increase in backups when moving from TU77s to TU78s. Our backups were done as two-drive backups, and that made things a LOT faster -- while one drive was rewinding, the other drive was spinning -- than normal, one-drive backups. We never tried running dual backups on two CPUs at the same time. I suspect that the HSC50 would have no trouble handling such a load. The TU78s reduce operations time significantly, and make the tape library a lot easier to deal with, but don't seem to run much faster due to the CPU bottleneck. I can't say for sure, but I would guess that you would find a TU78 on a VAX 750 to run at about the same speed as your TU80s. We don't use the TU80s for BACKUP, except in emergency, but when we have, they seem to run about 20 - 30% slower than the TU77s. Here are some representative times: Four RA81s -- BACKUP incremental on 3, full on 1: BACKUP/BUF=65534/BLOCK=5 with CRC TU77s: 2:30, 2:30, 2:15, 1:55, 2:00, 2:35, 1:55, 2:05, 2:25 TU78s: 1:45, 2:30, 2:15, 2:10, 3:00, 2:10 We will install an RH on an 8600 and put the drives on that next week; I'll send out results if there is a significant improvement. +-------------------------------+ | Joel M Snyder | BITNET: jms@arizmis.BITNET | Univ of Arizona Dep't of MIS | Internet: jms@mrsvax.mis.arizona.edu | Tucson, Arizona 85721 | Pseudo-PhoneNET: (602) 621-2748 +-------------------------------+ ICBM: 32 13 N / 110 58 W (I have gotten into trouble too many times to put any faith in disclaimers) "There's nothing here that an overdose of Seconal won't cure."