<font size=2 face="sans-serif">Hi, </font><br><font size=2 face="sans-serif">a simple approach is to use enhanced
dstat statistics on the NSD server side .. </font><br><br><font size=2 face="sans-serif">example:</font><br><font size=2 face="sans-serif">cp /usr/lpp/mmfs/samples/util/dstat_gpfsops.py.dstat.0.7
/usr/share/dstat/dstat_gpfsops.py</font><br><font size=2 face="sans-serif">export DSTAT_GPFS_WHAT="vio,vflush
-c -n -d -M gpfsops --nocolor"</font><br><br><font size=2 face="sans-serif">> dstat --gpfsops</font><br><br><font size=2 face="sans-serif">better and more fully, what you want...
configure ZIMON/ perfmon</font><br><font size=2 face="sans-serif">this gerates statistics like this.</font><br><font size=1 face="Courier">mmperfmon query compareNodes gpfs_nsdds_bytes_written
10</font><br><br><font size=1 face="Courier">Legend:</font><br><font size=1 face="Courier"> 1: gss01.frozen|GPFSNSDDisk|c1_LDATA1core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 2: gss01.frozen|GPFSNSDDisk|c1_LDATA1fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 3: gss01.frozen|GPFSNSDDisk|c1_LDATA2core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 4: gss01.frozen|GPFSNSDDisk|c1_LDATA2fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 5: gss01.frozen|GPFSNSDDisk|c1_LMETA1core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 6: gss01.frozen|GPFSNSDDisk|c1_LMETA1fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 7: gss01.frozen|GPFSNSDDisk|c1_LMETA2core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 8: gss01.frozen|GPFSNSDDisk|c1_LMETA2fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> 9: gss02.frozen|GPFSNSDDisk|c1_RDATA1core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">10: gss02.frozen|GPFSNSDDisk|c1_RDATA1fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">11: gss02.frozen|GPFSNSDDisk|c1_RDATA2core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">12: gss02.frozen|GPFSNSDDisk|c1_RDATA2fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">13: gss02.frozen|GPFSNSDDisk|c1_RMETA1core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">14: gss02.frozen|GPFSNSDDisk|c1_RMETA1fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">15: gss02.frozen|GPFSNSDDisk|c1_RMETA2core|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier">16: gss02.frozen|GPFSNSDDisk|c1_RMETA2fs1|gpfs_nsdds_bytes_written</font><br><font size=1 face="Courier"> </font><br><font size=1 face="Courier">Row
Timestamp gss01 gss01 gss01 gss01 gss01 gss01
gss01 gss01 gss02 gss02 gss02 gss02 gss02 gss02
gss02 gss02 </font><br><font size=1 face="Courier"> 1 2016-02-27-01:22:06
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 2 2016-02-27-01:22:07
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 3 2016-02-27-01:22:08
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 4 2016-02-27-01:22:09
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 5 2016-02-27-01:22:10
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 6 2016-02-27-01:22:11
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><font size=1 face="Courier"> 7 2016-02-27-01:22:12
0 83886080 0 67108864 0
0 0 0 0 16777216
0 16777216 0 0
0 0 </font><br><font size=1 face="Courier"> 8 2016-02-27-01:22:13
0 436207616 0 452984832 0 0
0 0 0 436207616 0 419430400
0 0 0 0 </font><br><font size=1 face="Courier"> 9 2016-02-27-01:22:14
0 16777216 0 0
0 0 0 0 0 67108864
0 83886080 0 0
0 0 </font><br><font size=1 face="Courier"> 10 2016-02-27-01:22:15
0 0 0
0 0 0 0 0
0 0 0
0 0 0 0
0 </font><br><br><br><font size=2 face="sans-serif">you can filter the overall IO's by filesystem
, or dedicated to some nodes.. it is very very flexible </font><br><font size=2 face="sans-serif">e.g. mmperfmon query compareNodes
gpfs_fs_bytes_written,gpfs_fs_bytes_read -n 5 -b 30 --filter gpfs_fs_name=beer
... and so on </font><br><br><font size=2 face="sans-serif">you may need some minutes to set it
up .. but once it is configured, it is very powerful ... </font><br><br><font size=2 face="sans-serif">have fun.. ;-)</font><br><br><br><br><br><font size=2 face="sans-serif"> </font><br><div><font size=2 face="sans-serif">Mit freundlichen Grüßen / Kind regards</font><br><br><font size=2 face="sans-serif"> <br>Olaf Weiser<br> <br>EMEA Storage Competence Center Mainz, German / IBM Systems, Storage Platform,<br>-------------------------------------------------------------------------------------------------------------------------------------------<br>IBM Deutschland<br>IBM Allee 1<br>71139 Ehningen<br>Phone: +49-170-579-44-66<br>E-Mail: olaf.weiser@de.ibm.com<br>-------------------------------------------------------------------------------------------------------------------------------------------<br>IBM Deutschland GmbH / Vorsitzender des Aufsichtsrats: Martin Jetter<br>Geschäftsführung: Martina Koederitz (Vorsitzende), Susanne Peter, Norbert
Janzen, Dr. Christian Keller, Ivo Koerner, Markus Koerner<br>Sitz der Gesellschaft: Ehningen / Registergericht: Amtsgericht Stuttgart,
HRB 14562 / WEEE-Reg.-Nr. DE 99369940 </font><br><br><br><br><font size=1 color=#5f5f5f face="sans-serif">From:
</font><font size=1 face="sans-serif">Brian Marshall <mimarsh2@vt.edu></font><br><font size=1 color=#5f5f5f face="sans-serif">To:
</font><font size=1 face="sans-serif">gpfsug-discuss@spectrumscale.org</font><br><font size=1 color=#5f5f5f face="sans-serif">Date:
</font><font size=1 face="sans-serif">07/12/2016 03:13 PM</font><br><font size=1 color=#5f5f5f face="sans-serif">Subject:
</font><font size=1 face="sans-serif">[gpfsug-discuss]
Aggregating filesystem performance</font><br><font size=1 color=#5f5f5f face="sans-serif">Sent by:
</font><font size=1 face="sans-serif">gpfsug-discuss-bounces@spectrumscale.org</font><br><hr noshade><br><br><br><font size=3>All,</font><br><br><font size=3>I have a Spectrum Scale 4.1 cluster serving data to 4
different client clusters (~800 client nodes total). I am looking
for ways to monitor filesystem performance to uncover network bottlenecks
or job usage patterns affecting performance.</font><br><br><font size=3>I received this info below from an IBM person. Does
anyone have examples of aggregating mmperfmon data? Is anyone doing
something different?</font><br><br><font size=3>"</font><font size=2 color=#2f2f2f face="Arial">mmpmon
does not currently aggregate cluster-wide data. As of SS 4.1.x you can
look at "mmperfmon query" as well, but it also primarily only
provides node specific data. The tools are built to script performance
data but there aren't any current scripts available for you to use within
SS (except for what might be on the SS wiki page). It would likely be something
you guys would need to build, that's what other clients have done.</font><font size=3>"</font><br><br><br><font size=3>Thank you,</font><br><font size=3>Brian Marshall</font><br><font size=3>Virginia Tech - Advanced Research Computing</font><tt><font size=2>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=2>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=2><br></font></tt><br></div><BR>