Re: [Networker] Solaris 9, Exabyte M2, and ST_BUFFERED_WRITES
2004-03-02 16:31:57
> Anyone have any idea what this implication of not using
> ST_BUFFERED_WRITES with M2 drives?
Knowing nothing in particular about the M2, I would suspect that there
would be an increase in latency and possibly a slight drop in
throughput. Now the drive is not allowed to acknowledge a write as
being complete until it is actually on the tape, not just the buffer.
If you're operating with lots of data (streaming), this may not affect
you much.
If the drive really didn't support it properly, it's possible that it
could acknowlege a write near the end of the tape that couldn't actually
fit on the tape. Networker checks the last blocks written at the end of
the tape by default, so at least it could determine if corruption had
occurred by that method before completing the saveset.
I'm not certain how Solaris has determined that that particular drive
does not support it.
--
Darren Dunham ddunham AT taos DOT com
Senior Technical Consultant TAOS http://www.taos.com/
Got some Dr Pepper? San Francisco, CA bay area
< This line left intentionally blank to confuse you. >
--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
Previous by Date: |
[Networker] Solaris 9, Exabyte M2, and ST_BUFFERED_WRITES, Matthew Huff |
Next by Date: |
Re: [Networker] Solaris 9, Exabyte M2, and ST_BUFFERED_WRITES, Itzik Meirson |
Previous by Thread: |
[Networker] Solaris 9, Exabyte M2, and ST_BUFFERED_WRITES, Matthew Huff |
Next by Thread: |
Re: [Networker] Solaris 9, Exabyte M2, and ST_BUFFERED_WRITES, Itzik Meirson |
Indexes: |
[Date]
[Thread]
[Top]
[All Lists] |
|
|