RE: Determine bottlenecks during backup

> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx 
> [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Jones, John (US)
> Sent: Thursday, February 23, 2006 11:35 AM
> To: Midrange Systems Technical Discussion
> Subject: RE: Determine bottlenecks during backup
> 
> What kind of tape drive?

It's actually a virtual tape subsystem from Cybernetics.  The iSeries
sees it as a SCSI attached 3580, but the data is actually going to a
RAID array of 400GB SATA drives.

> 
> The SCSI interface in this case is very likely not the 
> bottleneck; I've
> got 5702s that are doing about 275MB/s (compressed) with LTO3 drives.
> That leaves the potential bottlenecks as:
> 
> 1. Tape drive.  If LTO2 or LTO3, this should not be the bottleneck
> unless the drive is malfunctioning or dirty.

According to the specs, the virtual tape drive should be able to handle
200MB/s

> 2. System bus.  Could be the problem.  Where is bus 28?  If not in the
> CEC, what type of tower?  I had LTO2 drives on our former 
> 830; one on a
> controller in the CEC and one in a migration tower (a former 
> 720 frame).
> The drive on the controller in the CEC was significantly faster (more
> than 2x) than the drive in the tower even though the drives themselves
> were identical.

Bus 28 is in a 5088 sitting on top of a 5094.

> 3. System (CPU/RAM).  This is probably not the bottleneck unless the
> partition is RAM-constrained.

System has 4GB RAM, the partition in question is using 3GB.

> 4. Disk subsystem.

I believe this is covered in the original post.

> 5. Object type.  Fewer-but-larger objects will save faster 
> than lots of
> small objects.  The more objects there are the higher the system
> overhead.

127,426 objects average size 1,230,907 bytes  (excluding one >
9,999,999,999)

(Why can't I find a median function in SQL?)

> 6. Save type.  Turn on compression for the non-SAVSYS pieces; 
> you've got
> plenty of CPU to handle it.

I'll look at this also.

> 
> Also, try to figure out which part of the 21 is taking a lot of time.
> Is it IFS, NONSYS, SAVSYS, etc.?  Easiest way is to look at 
> the job log
> afterwards and look at the timestamps.

SAVSYS: 9min
SAVLIB *NONSYS: 64 min
SAVDLO: 1min
SAV: 15min

> 
> John A. Jones, CISSP
> Americas Information Security Officer
> Jones Lang LaSalle, Inc.
> V: +1-630-455-2787 F: +1-312-601-1782
> john.jones@xxxxxxxxxx
> 

Thanks for the help,
Charles Wilt



> -----Original Message-----
> From: midrange-l-bounces@xxxxxxxxxxxx
> [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Wilt, Charles
> Sent: Thursday, February 23, 2006 10:02 AM
> To: Midrange Systems Technical Discussion
> Subject: Determine bottlenecks during backup
> 
> All,
> 
> I'm trying to evaluate the performance of our new backup system.
> 
> The tape device is rated at 200 MB/s and is attached to an 
> 160MB/s 5702
> SCSI card.
> 
> However, on my first test save menu option 21 save I saved 238.7 GB of
> data (uncompressed) in about 90 minutes, a throughput of 
> about 45 MB/s.
> 
> The iSeries is a model 810 with 2700CPW of which 87% (2349CPW) is
> assigned to the primary partition being backed up.
> 
> System bus 1 has a 2757 Controller card with 12x4326 RAIDed.  Of those
> 12 disks, the system ASP contains 10 disks, and a user ASP is 
> configured
> with 2 disks.
> 
> System bus 28 has the 5702 controller card.
> 
> 
> How can I tell if the save is bottlenecked by CPU, disk, bus, or tape
> device?
> 
> Thanks,
> 
> Charles Wilt
> --
> iSeries Systems Administrator / Developer Mitsubishi Electric 
> Automotive
> America
> ph: 513-573-4343
> fax: 513-398-1121
>  
> 
> --
> This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
> list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
> unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, 
> please take a
> moment to review the archives at 
> http://archive.midrange.com/midrange-l.
> 
> 
> 
> This email is for the use of the intended recipient(s) only.  
> If you have received this email in error, please notify the 
> sender immediately and then delete it.  If you are not the 
> intended recipient, you must not keep, use, disclose, copy or 
> distribute this email without the author's prior permission.  
> We have taken precautions to minimize the risk of 
> transmitting software viruses, but we advise you to carry out 
> your own virus checks on any attachment to this message.  We 
> cannot accept liability for any loss or damage caused by 
> software viruses.  The information contained in this 
> communication may be confidential and may be subject to the 
> attorney-client privilege. If you are the intended recipient 
> and you do not wish to receive similar electronic messages 
> from us in the future then please respond to the sender to 
> this effect.
> 
> -- 
> This is the Midrange Systems Technical Discussion 
> (MIDRANGE-L) mailing list
> To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> or email: MIDRANGE-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/midrange-l.
> 
>