Russell Adams <RLAdams AT kelsey-seybold DOT com>
Sent by: 
owner-amanda-users AT amanda DOT org
04/24/2003 10:26 AM

>> Of course hardware compression has one big advantage: speed.  If you
>> have to fit lots of data in a small backup window at night, and your 
>> cpu's are not fast enough, or interfere with the nightly cpu-intensive
>> jobs, then hardware compression is the way to go.
>
>
>I have to take issue with this. My experience with hardware
>compression on DDS and DLT drives is that when enabled it makes things
>incredibly slow.
>
>My DLT IV (er, 7000?) drives can do 4.9 MB/s uncompressed, and only
>800KB/s in compressed mode.
>
>I have to seriously believe that the hardware compression ratio could
>easily be surpassed by a modestly fast host performing software
>compression. Also with Amanda doing it, it'll already be compressed on
>the holding disk before streaming to tape.
>
>Yet another reason I use client side compression.

I've used hard comp all along and I had the exact opposite results when I 
recently tested soft comp. My test times more than doubled & the cpu use 
went up. I have a somewhat  of a busy environment (CCase, Oracle, multiple 
massive builds, Win2k, samba, email,  web ~ 350GB total) to cover and 
can't afford the hit soft comp gives out. I'm using AIT-1 & ATI-2 drives.


--
toby bluhm
philips medical systems, it support, mr development, cleveland ohio
tobias.bluhm AT philips DOT com
440-483-5323









Russell Adams <RLAdams AT kelsey-seybold DOT com>
Sent by: 
owner-amanda-users AT amanda DOT org
04/24/2003 10:26 AM

 
        To:     amanda-users AT amanda DOT org
        cc:     (bcc: Tobias Bluhm/CLE/MS/PHILIPS)
        Subject:        Re: hardware vs software compression (was Re: 
amflush/amcheck not in 
sync?)
        Classification: 



> Of course hardware compression has one big advantage: speed.  If you
> have to fit lots of data in a small backup window at night, and your 
> cpu's are not fast enough, or interfere with the nightly cpu-intensive
> jobs, then hardware compression is the way to go.


I have to take issue with this. My experience with hardware
compression on DDS and DLT drives is that when enabled it makes things
incredibly slow.

My DLT IV (er, 7000?) drives can do 4.9 MB/s uncompressed, and only
800KB/s in compressed mode.

I have to seriously believe that the hardware compression ratio could
easily be surpassed by a modestly fast host performing software
compression. Also with Amanda doing it, it'll already be compressed on
the holding disk before streaming to tape.

Yet another reason I use client side compression.

Russell