Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 13 Nov 1998 15:27:47 -0800 (PST)
From:      Matthew Dillon <dillon@apollo.backplane.com>
To:        dag-erli@ifi.uio.no (Dag-Erling C. =?iso-8859-1?Q?Sm=F8rgrav?= )
Cc:        Marc Slemko <marcs@znep.com>, David Wolfskill <dhw@whistle.com>, hackers@FreeBSD.ORG
Subject:   Re: dump(8) very slow
Message-ID:  <199811132327.PAA06232@apollo.backplane.com>
References:  <Pine.BSF.4.05.9811130909500.12077-100000@alive.znep.com> <199811131747.JAA05198@apollo.backplane.com> <xzp4ss3cptx.fsf@hrotti.ifi.uio.no>

next in thread | previous in thread | raw e-mail | index | archive | help

:Matthew Dillon <dillon@apollo.backplane.com> writes:
:>                                                 I gzip -2 the dump output on
:>     the remote machine before it comes back through the ssh pipe.
:
:Why not turn on compression in ssh (-C) rather than add an extra stage
:to the pipeline?

   I am saving gzip'd dumps to tape directly.  I know, I know, it means you'd
   better not have a tape error in a dump.  Too bad.  Exabytes do duel-head
   verification anyway, though you definitely have to clean the drives twice
   a week (at the volume levels we push through the drives) to maintain 
   optimal performance (which is 3 MBytes/sec for an exabyte Mammoth).

   The backup machine is already cpu-limited doing the encryption (even though
   the network is switched and secured, we don't trust it).  If I had the
   backup machine do compression (ssh -C) as well, or if I had it gunzip
   the remotely-gzip'd dump, the 2.5 MBytes/sec would because 1.5 MBytes/sec.
   If I have the tape drive do the compression, I get sub-optimal capacity 
   verses pushing gzip'd dumps to a tape drive in uncompressed mode.

   Sometimes I think about turning off ssh's link encryption, but then I think 
   a little more and say 'nahh, better to leave it on'.  Once we upgrade the
   machine we'll be able to dump 6 MBytes/sec over the network, which matches
   two 3 MByte/sec tape drives very well.

						-Matt

:DES
:-- 
:Dag-Erling Smørgrav - dag-erli@ifi.uio.no
:
:To Unsubscribe: send mail to majordomo@FreeBSD.org
:with "unsubscribe freebsd-hackers" in the body of the message
:

    Matthew Dillon  Engineering, HiWay Technologies, Inc. & BEST Internet 
                    Communications & God knows what else.
    <dillon@backplane.com> (Please include original email in any response)    

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811132327.PAA06232>