There is _no_ mistake in the math. There are (most likely) multiple streams going on. 
I have asked about this before and think I have it straighten out.

Look at the aggregate and network rates. You probably have 4 streams at 10 MB/s each, 
for a total of 40 MB/s.

Try using 'run q_ses_stats' (and q ses) during the backup. It will show any multiple 
streams. Look in the SUMMARY table to see the stats after the backup has run.

This behavior is consistent with what I see for a small number (535) of large files 
(in 113 GB). Here's a good example where the aggregate exceeds the network:

04/25/02   03:19:55      ANE4952I (Session: 1994, Node: UNXR)  Total number of      
objects inspected:   81,678                              
04/25/02   03:19:55      ANE4954I (Session: 1994, Node: UNXR)  Total number of      
objects backed up:      665                              
04/25/02   03:19:55      ANE4958I (Session: 1994, Node: UNXR)  Total number of      
objects updated:          0                              
04/25/02   03:19:55      ANE4960I (Session: 1994, Node: UNXR)  Total number of      
objects rebound:          0                              
04/25/02   03:19:55      ANE4957I (Session: 1994, Node: UNXR)  Total number of      
objects deleted:          0                              
04/25/02   03:19:55      ANE4970I (Session: 1994, Node: UNXR)  Total number of      
objects expired:        558                              
04/25/02   03:19:55      ANE4959I (Session: 1994, Node: UNXR)  Total number of      
objects failed:           0                              
04/25/02   03:19:55      ANE4961I (Session: 1994, Node: UNXR)  Total number of      
bytes transferred:    54.93 GB                           
04/25/02   03:19:55      ANE4963I (Session: 1994, Node: UNXR)  Data transfer time:   
3,067.39 sec                               
04/25/02   03:19:55      ANE4966I (Session: 1994, Node: UNXR)  Network data            
transfer rate:        18,780.88 KB/sec                   
04/25/02   03:19:55      ANE4967I (Session: 1994, Node: UNXR)  Aggregate data       
transfer rate:      27,531.04 KB/sec                     
04/25/02   03:19:55      ANE4968I (Session: 1994, Node: UNXR)  Objects compressed      
 by:                    0%                                
04/25/02   03:19:55      ANE4964I (Session: 1994, Node: UNXR)  Elapsed processing     
time:            00:34:52                                

Miles


----------------------------------------------------------------------------------------------
Miles Purdy 
System Manager
Farm Income Programs Directorate
Winnipeg, MB, CA
[EMAIL PROTECTED]
ph: (204) 984-1602 fax: (204) 983-7557

"If you hold a UNIX shell up to your ear, can you hear the C?"
-------------------------------------------------------------------------------------------------

>>> [EMAIL PROTECTED] 25-Apr-02 5:11:07 AM >>>
I guess we al know TSM F***s up the math big time. I have seen percetages of
over 150% for pct idle/comm/mediaw in q node output....

On Wed, 24 Apr 2002 17:42:49 -0500
"Bill Mansfield" <[EMAIL PROTECTED]> wrote:

> Also, look at aggregate transfer rate: 43 GB/Sec is pretty good.
>
> _____________________________
> William Mansfield
> Senior Consultant
> Solution Technology, Inc
>
>
>
>
>
> "Ochs, Duane" <[EMAIL PROTECTED]>
> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]>
> 04/24/2002 03:19 PM
> Please respond to "ADSM: Dist Stor Manager"
>
>
>         To:     [EMAIL PROTECTED] 
>         cc:
>         Subject:        Has anybody else seen anything like this ?
>
>
> Process starts with session # 51170 @ 10:35 and finishes @ 14:28. But it
> reports 201 hours in elapsed time. I have looked at the logs for the last
> two weeks and have not seen anything like this.
> TSM Client 4.2.1.26
> NT 4.0 SP5
>
> 04/22/02 10:35:17 ANR0406I Session 51170 started for node NODE1 (WinNT)
> 04/22/02 14:28:47 ANE4952I (Session: 51170, Node: NODE1) Total number of
> objects inspected: 539
> 04/22/02 14:28:47 ANE4953I (Session: 51170, Node: NODE1) Total number of
> objects archived: 525
> 04/22/02 14:28:47 ANE4958I (Session: 51170, Node: NODE1) Total number of
> objects updated: 0
> 04/22/02 14:28:47 ANE4960I (Session: 51170, Node: NODE1) Total number of
> objects rebound: 0
> 04/22/02 14:28:47 ANE4957I (Session: 51170, Node: NODE1) Total number of
> objects deleted: 481
> 04/22/02 14:28:47 ANE4970I (Session: 51170, Node: NODE1) Total number of
> objects expired: 0
> 04/22/02 14:28:47 ANE4959I (Session: 51170, Node: NODE1) Total number of
> objects failed: 0
> 04/22/02 14:28:47 ANE4961I (Session: 51170, Node: NODE1) Total number of
> bytes transferred: 113.79 GB
> 04/22/02 14:28:47 ANE4963I (Session: 51170, Node: NODE1) Data transfer
> time:
> 11,558.91 sec
> 04/22/02 14:28:47 ANE4966I (Session: 51170, Node: NODE1) Network data
> transfer rate: 10,323.13 KB/sec
> 04/22/02 14:28:47 ANE4967I (Session: 51170, Node: NODE1) Aggregate data
> transfer rate: 42,949,643.92 KB/sec
> 04/22/02 14:28:47 ANE4968I (Session: 51170, Node: NODE1) Objects
> compressed
> by: 0%
> 04/22/02 14:28:47 ANE4964I (Session: 51170, Node: NODE1) Elapsed
> processing
> time: 201:18:58
> 04/22/02 14:28:47 ANR0403I Session 51170 ended for node NODE1 (WinNT).
>
>
> Duane Ochs
> Systems Administration
> Quad/Graphics Inc.
> 414.566.2375


--
Met vriendelijke groeten,

Remco Post

SARA - Stichting Academisch Rekencentrum Amsterdam    http://www.sara.nl 
High Performance Computing  Tel. +31 20 592 8008    Fax. +31 20 668 3167

"I really didn't foresee the Internet. But then, neither did the computer
industry. Not that that tells us very much of course - the computer industry
didn't even foresee that the century was going to end." -- Douglas Adams

Reply via email to