How do i view a CX3xx's CPU usage?

There are two methods to monitoring performance on a CLARiiON array:

1. if your array is running flare release 24 or higher, there is a default performance tool installed called Navisphere Analyzer. You can use this tool to gather performance statistics into an archive. In the default mode, the archives can only be analyzed by CLARiiON Support. If you have the licensed version of Analyzer you can open and view these archives on your workstation

2. If you have the licensed version of Analyzer installed, you can run a real-time version of Analyzer - it will display a graph on your workstation from within Navisphere - just right click on an object such as as SP or a LUN and select Analyzer/Performance Detail - after a couple of minutes, you should start to see some data points. These are updated every 60 seconds.

In order for both of the above methods to work, you must have "started" the data logging function. Go to Tools/Analyzer/Data Logging - you'll see the START button at the bottom of the screen. This is also where you can tell if your array has the licensed Analyzer installed. If the "Stop Automatically After" button is grayed out, you do not have the licensed Analyzer or the monitor is running.

If you are running an older version of Flare, release 22 or earlier, most of the above is the same except for the START button - in earlier releases you needed to enable "Statistics Logging" from the General tab on the array Properties screen.

You should also obtain the below documents for an explaination of Best Practices for CLARiiON arrays.


EMC CLARiiON Best Practices for Fibre Channel Storage: FLARE Release 26 Firmware Update - Best Practices Planning

http://powerlink.emc.com/km/live1/en_US/Offering_Technical/White_Paper/H2358_clariion_best_prac_fibre_chnl_wp_ldv.pdf

EMC CLARiiON Fibre Channel Storage Fundamentals - Technology Concepts and Business Considerations

http://powerlink.emc.com/km/live1/en_US/Offering_Technical/White_Paper/H1049_emc_clariion_fibre_channel_storage_fundamentals_ldv.pdf


regards,

glen kelley

Thank you!

one method is to log on to storage processor directly but this is not allowed for customer so will not share the details

i'm pretty sure you are right, but still....it would be nice if i can log into the array/SP and look at real time performance and a breakdown of CPU circles, i would really appreciated if you guys can help me out.

Another point about using the SP directly is that the information provided would not really provide you with any useful data - you want to look at LUN and Disks performance - this is provided in the Analyzer archives, not from the Task Manager.

glen

that is assuming LUN activity is the only thing that's consuming CPU circles, i had a problem recently that the CPU spiked above 50% and stopped an online FLARE upgrade, and i didn't think there was massive IO from hosts end, is it possible other stuff such as mirroring would consume a large part of CPU circles?

That is engineering only access from EMC service personnel. No end users have access to that information. Kiran really should not have mentioned it as it is not a realistic option.

And those would still be items that are monitored from within Navisphere.

There are lots of array activities that can affect the performance - mirroring, SNAPclones, SANcopy all affect performance - it's more difficult to track these down (especially mirroring as the the other array may be the cause or the latency on the link). But these numbers are collected in the analyzer (thanks AranH).

Please see knowlegebase article emc176825 for additional information.


glen

Message was edited by:
kelleg

Message was edited by:
kelleg

i did mention about logging onto embedded windows server on SP

however, it is not the correct way to gauge performance. you can only see a collective cpu utilisation by flare code running on that processor. glen's comments stand valid otherwise...

ok, i guess you guys are not going to help me out then

just kidding, i have verified my array does have Analyzer licensed, and will read up on the docs suggested, thanks so much for everybody's time and help!

except for not letting me know the "Administrator" account info, Kiiiiiiiidding

Kiran is a EMC partner I believe, which is why he has access to the "secret" passwords End users of storage like you and me don't have access to that information and don't need it. There is no need for an end user to ever be mucking around in the storage processors.

Sorry you got side-tracked on this, but Analyzer is the only performance monitoring tool you need for the clariion.

Let me know if you have any more questions about using navisphere analyzer. You will find that it is very useful and has a ton of information available for analysis. I recommend setting up a batch script that runs everyday and collects the analyzer files from the array. This way you have historical data of your array for trending and analysis.

Good luck,
Aran

Remeber that in release 24 and up, you can enable Periodic Archive and not set the "Stop After" and the array will automatically create the NARs for you. You do have to remember to pull them off the array (the script is easier in this regard) - the cx3 series will hold about 200MB of NAR data before it starts to overwrite.

glen