Debbie,

Below a posting from a few weeks ago.


LTO Microcode - Perf Improvement
From: arhoads <arhoads AT PACBELL DOT NET>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Mon, 12 Mar 2001 06:10:46 -0800


I'm posting this to the ADSM-L list so that anyone to whom it applies
can gain the benefit ASAP.  I get it from an IBM internal/Business
Partner mailing and recieved permission to post it here.

Regards,  Steffan

+++++++++++

The latest LTO drive microcode (0CE1) provides a
significant performance improvement for customers doing
certain types of tape operations.  This level of code
dramatically improves the overheads associated with
buffer flush (write 0 filemarks) and end of file (write
2 filemarks, backspace 1 filemark).

TSM does a buffer flush after every transaction.  The
end of file impact would be greatest for customers
writing a series of files to the LTO tape drive.  For
tape applications that intensively use buffer flush or
are writing a series of files we have seen 30% to 50%
elapsed time improvements after installing 0CE1.  0BN1
was the previous LTO drive microcode release.

For 3580, 3581, and 3583 mach types (Customer Setup),
the customer can download 0CE1 from:

     ftp://service.boulder.ibm.com/storage/358x

     Choose 0CE1.fmr

Instructions for verifying the current LTO drive
microcode release and how to install the new release
can be found in the IBM Ultrium Device Drivers
Installation and Users Guide available from:

     ftp://ftp.software.ibm.com/storage/devdrvr/

For 3584s, have the customer request his CE update the
LTO drive microcode.




On Friday 20 April 2001 19:19, you wrote:
> I have not been monitoring the list for a while, so forgive me if I ask
> something that has been covered previously...
>
> We recently upgraded our 3466 model A01 with two 3575 tape libraries to a
> 3466 model C01 with a 3584 tape library.  We also upgraded from ADSM 3.1.1
> to TSM 4.1.  I have begun to notice that our tape operations are taking
> significantly longer than they did before.  Simple restores take forever
> between the mount and the start of data transfer, the time it takes to
> backup our DB has tripled (and the DB is smaller), and restores of large
> Oracle and Exchange databases are taking about twice as long.  The process
> of backing data up to the copypool for offsite storage is also taking a
> very long time.  I have a process running now that has backed up 21Gb from
> one tape in the primary pool to one tape in the copypool, in other words a
> straight tape to tape copy, over a two hour time period.
>
> Does anyone out there have any similar experience?  We do have support from
> both Tivoli and IBM hardware support, any suggestions on who to call first?
>
> Debbie Weeks
> University of South Florida
> Information Technologies/Technical Support
> ********
> *Debbie AT ADMIN.USF DOT EDU
> *(813)974-6926*S/C 574-6926
> Fax (813) 974-3054

-- 
With regards / Mit freundlichen Grüßen
With regards / Mit freundlichen Grüßen

    Leopold Hameder

 |  Heinrich Bauer Produktions KG PSC      |  email: lhameder AT hbv DOT de    |
 |  Abteilung Server & Concepts            |  phone: 040 - 3019 - 5675  |
 |  Burchardstr. 11, 20077 Hamburg         |  fax:   040 - 3019 - 5668  |