[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference stkhlm::magtape

Title:MAGNETIC TAPEDRIVES
Moderator:STKHLM::GJOHNSSON
Created:Mon Sep 21 1987
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:3775
Total number of notes:13147

2114.0. "TZ87 access times!?" by MORLEY::RUSHWORTH (I'm the missing BOD!) Wed Mar 09 1994 21:10

T.RTitleUserPersonal
Name
DateLines
2114.1It dependsELWOOD::KAPLANLarry Kaplan, DTN: 237-6872Wed Mar 09 1994 21:3018
2114.2SUN to be specific...BAHTAT::RUSHWORTHWhy is there only 1 Monopolies Commission?Thu Mar 10 1994 18:347
2114.3ELWOOD::KAPLANLarry Kaplan, DTN: 237-6872Thu Mar 10 1994 20:593
2114.4Customer QuestionODIXIE::KEMPWed Mar 16 1994 20:4213
2114.5ELWOOD::PETERSWed Mar 16 1994 23:2011
2114.6ThanksODIXIE::KEMPThu Mar 17 1994 00:557
2114.7More questions...MILBRN::RUSHWORTHWhy is there only 1 Monopolies Commission?Mon Mar 21 1994 03:0521
2114.8ELWOOD::PETERSMon Mar 21 1994 19:0117
2114.9SSDEVO::FROEHLINBetween my ears? 80% water or what?Mon Mar 21 1994 21:4613
2114.10How long for a full restorationOTOOA::LAVIGNEWed May 11 1994 02:2111
2114.11DECWET::GETSINGERProd Mgr - POLYCNTR HSM for UNIXThu Jun 09 1994 11:3317
2114.12OSF I/O guru neededELWOOD::KAPLANLarry Kaplan, DTN: 237-6872Thu Jun 09 1994 20:3115
2114.13DECWET::GETSINGERProd Mgr - POLYCNTR HSM for UNIXFri Jun 10 1994 21:342
2114.14at long last...CUJO::SAMPSONThu Jan 09 1997 08:5819
2114.15TAPE::PETERSFri Jan 10 1997 00:319
2114.16Faster skipfile is in V7.1STAR::S_SOMMERFri Jan 10 1997 19:3317
2114.17Be sure MKSET is used locally on 7.1 tape dens setEVMS::EVERHARTWed Jan 29 1997 18:289
    One thing to note about the new MK skipfiles stuff. If using MKSET to
    alter tape behavior defaults, remember it must be run on the same
    machine where the tape is on a local SCSI bus; mkset does not work
    across servers. (Rotorooting the TMSCP spec was not feasible.)
    
    Thus if say "MKA300:" doesn't work but you'd use "$3$MKA300:", the
    device is probably not local...
    glenn
    
2114.18Customer has compromised the new algorithm?CSC32::D_BROWNDave Brown CSC-VSG/INTDRVTue Jun 03 1997 01:2021
    
    Could the new algorithm at all be responsible for a dramatic *increase*
    in backup time to a TZ87 (TZ887)? I'm working with a customer who since
    upgrading to V7.1 Alpha has noticed backups sometimes take up to 10
    hours longer than they used to. His procedure is quite poor but used to
    not take so long and I'm wondering if performing backups in such a
    manner will kill the efficiency of the new algorithm. Here's what he 
    does for each disk volume; one batch job per backup:
    
    	1) Mount tape
    	2) Backup disk
    	3) Dismount tape
        4) Next batch job starts with step 1
    
    This is repeated until all his disks have been backed up.
    	
    
    Thanks,
    
    Dave
    	
2114.19COOKIE::FROEHLINVMS...riding into the setting sun!Wed Jun 04 1997 00:2317
.18>upgrading to V7.1 Alpha has noticed backups sometimes take up to 10
    
    Sometimes? Exactly when? Could it be that sometimes it needs a
    continutaion tape and there's no one around for 10 hours to mount a new
    one?
    
.18>	1) Mount tape
.18>	2) Backup disk
.18>  	3) Dismount tape
.18>    4) Next batch job starts with step 1
    
    BACKUP will do a "SKIP TMs 32000" to reach the logical-end-of-tape to
    append another save set. I would expect the new behavior in MKDRIVER
    to improve the skip time. What kind of SCSI controller is this tape
    connected to?
    
    Guenther