What tweaks did you do?
Yes, should of mentioned the server is dedicated for this purpose and is
reasonably resourced:
-dedicated SAS card for tape drives
-64GB RAM
-2x Xeon E5-2640 v3, so plenty of cores
-dedicated holding disk areas (RAID6) .. two of these coming in at approx.
32.7TB, which will match up to LTO8 quite nicely hopefully. Comprised 8x 6TB
disks.
By the sounds of it, not too dissimilar to your setup.
-------------
David Simpson - Senior Systems Engineer
ARCCA, Redwood Building,
King Edward VII Avenue,
Cardiff, CF10 3NB
David Simpson - peiriannydd uwch systemau
ARCCA, Adeilad Redwood,
King Edward VII Avenue,
Caerdydd, CF10 3NB
simpso...@cardiff.ac.uk
+44 29208 74657
-----Original Message-----
From: Chris Hoogendyk <hoogen...@bio.umass.edu>
Sent: 28 September 2021 20:00
To: David Simpson <simpso...@cardiff.ac.uk>; AMANDA users
<amanda-users@amanda.org>
Subject: Re: LTO8 tape profile - amtapetype
External email to Cardiff University - Take care when replying/opening
attachments or links.
Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth ateb/agor
atodiadau neu ddolenni.
You should get substantially faster. That said, throughput to the tape will
depend on a lot of things about your system and where the bottleneck actually
is. The maximum throughput of dual SAS2 is 6Gbit/s, which would be 750MB/s. So,
you aren't likely to come close to the 900MB/s for compressible data that LTO8
is rated for. If there are other things going on using the SAS bandwidth, like
your hard drive systems, then that bites into what's available for pushing out
to tape. Then there is the cpu for generating the data stream.
I've only gotten up to LTO7. However, it took some tweaking of my supermicro
servers to approach the rated speed for the LTO7. The tape drives are on a
dedicated SAS card. The disk drives are on a separate SAS card. My servers have
two cpus with I think 16 core each, and 64GB memory. So, even though they are
doing a lot of other stuff, Amanda can still push the LTO7 approaching its
rated speed.
On 9/28/21 5:10 AM, David Simpson wrote:
>
> Just ran amtapetype against an LTO8 tape + HP MSL3040 (with SAS drives).
> Debian 11.
>
> Some observations:
>
> -compression on, expected
> -length looks ok
>
> -speed .. should I have expected more? Seemed a bit low I thought..
>
> -no LEOM.. I don't know if it's the hardware or the Debian 11 kernel
> or both, though understand this is not essential
>
>
>
> Checking for FSF_AFTER_FILEMARK requirement
>
> Applying heuristic check for compression.
>
> Wrote random (uncompressible) data at 96137551.7377049 bytes/sec
>
> Wrote fixed (compressible) data at 167554018.742857 bytes/sec
>
> Compression: enabled
>
> Writing one file to fill the volume.
>
> Wrote 12011529469952 bytes at 94902 kb/sec
>
> Writing smaller files (120115265536 bytes) to determine filemark.
>
> define tapetype unknown-tapetype {
>
> comment "Created by amtapetype; compression enabled"
>
> length 11730009248 kbytes
>
> filemark 4132 kbytes
>
> speed 94902 kps
>
> blocksize 32 kbytes
>
> }
>
> # LEOM is not supported for this drive and kernel
>
>
>
>
>
> -------------
>
> David Simpson - Senior Systems Engineer
>
> ARCCA, Redwood Building,
>
> King Edward VII Avenue,
>
> Cardiff, CF10 3NB
>
> David Simpson - peiriannydd uwch systemau
>
> ARCCA, Adeilad Redwood,
>
> King Edward VII Avenue,
>
> Caerdydd, CF10 3NB
>
> simpso...@cardiff.ac.uk <mailto:simpso...@cardiff.ac.uk>
>
> +44 29208 74657
>
--
---------------
Chris Hoogendyk
-
O__ ---- Systems Administrator, Retired
c/ /'_ --- Biology & Geosciences Departments
(*) \(*) -- 315 Morrill Science Center III ~~~~~~~~~~ - University of
Massachusetts, Amherst
<hoogen...@bio.umass.edu>
---------------
Erdös 4
