Highlighted
2 Bronze

Re: Ask the Expert: Performance Calculations on Clariion/VNX

Hi Dynamox... We have recently upgraded to INYO 5.32. Does this feature really exists on this code... The cache intelligence for understanding the log (sequential pattern) and database and other Random pattern and allocating cache based on that?

0 Kudos
Highlighted
7 Thorium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

swadeey123 wrote:

Hi Dynamox... We have recently upgraded to INYO 5.32. Does this feature really exists on this code... The cache intelligence for understanding the log (sequential pattern) and database and other Random pattern and allocating cache based on that?

unfortunately i do not know the specifics but was told there were improvements. (not sure how one would quantify them in terms of iops)

0 Kudos
Highlighted
2 Bronze

Re: Ask the Expert: Performance Calculations on Clariion/VNX

Ok no problems, as we were in some trouble and this could have helped us, so wanted to check.

Guys can anyone of you help me with my current performance problem, maybe guide me in troubleshooting it. I will tell this setup is long back implemented and been played around a several time before i joined, here is how it is.

1. 3 storage boxes alocatting space to a SQL 2000 servers (to start the mess with all of different genes to add up this have got a long story associated with it)

2. OS is Windows 2003 virtual machine running on ESX Vcenter 4.1 having datastores from DMX, Clarrion, and VNX. 3. We are having performance issue and some time we do see huge read write latency with almost from all the three boxes Clariion showing the highest latency.

4. Another mess we have around 3 to 4 drives which have both "Database" and "Logs" going to them.
5. DMX lun (datastore for ESX) are R 5 (15 K drives), Clariion Lun is a datastore (only this server is running on this) of Radi 1/0 with 8 drives, VNX LUN is coming from a storage pool Raid 5 (mixed pool) with (Auot tiering and FAST Cache enabled)

6. What issue we have is one of our process which runs everyday have been taking considerably high time, which access this database (both read and write) IO types. We checked all the things at the ESX level and do not find any issues, we are in process of collection of "perfmon" for the server. We suspect the LUN coming from Clariion to be problematic (as that is the only lun that is not common without anything else hosting this host), For sure it will be difficult for any of you to suggest without having clear pricture but can some one guide if they have faced this similar issue. We are actually looking for immediate and maybe short term fix (as we might decommison this system soon but not very soon (2 to 6 months) Can someone guide... Please do not tell me how messy it is as i know 🙂

0 Kudos
Highlighted
2 Iron

Re: Ask the Expert: Performance Calculations on Clariion/VNX

I was curious so checked it out here http://www.emc.com/collateral/software/white-papers/h8018-fast-cache-oracle-wp.pdf

It is not enabled on any of the use cases here and in the conclusion is says to Disable FAST Cache on LUNS where Online Redo reside.

Cheers,

Victor

Highlighted
4 Beryllium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

FAST Cache should be disabled on the online redo logs. Logs are small block sequential writes; FAST Cache will not help

Highlighted
2 Iron

Re: Ask the Expert: Performance Calculations on Clariion/VNX

Yes that makes sense. I thought my reply would appear at the relevant area of the FAST cache discussion area but appeared at the end so my comment is out of context here so  apologies.

The discussion on this end of the discussion is the new VNX Snapshot technology. This is now ROW (Redirect on write) to new location within the same VNX pool and no longer COFW (Copy on First Write) holding writes to primary lun until original data copied to the reserve lun pool. This defininitely seems better for performance as the write/change to the primary can occur straight away and a read for example of the source lun does not need to be constructed/read from two different places.

0 Kudos
Highlighted
6 Indium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

I’m sure there’s something wrong with the layout. I’m seeing the same thing. Also previously I could view the whole discussion in 1 view and now there are 8 pages

Mark will fix it, I’m sure

0 Kudos
Highlighted
6 Indium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

Good one! Thanks for the addition!!

0 Kudos
Highlighted
6 Indium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

I'm wondering if logging is sequential or random I/O, because that might change things.

Anyone?

0 Kudos
Highlighted
7 Thorium

Re: Ask the Expert: Performance Calculations on Clariion/VNX

8.2.3.2 Redo Log Files

If the high-I/O files are redo log files, then consider splitting the redo log files from the other files. Possible configurations can include the following:

  • Placing all redo logs on one disk without any other files. Also consider availability; members of the same group should be on different physical disks and controllers for recoverability purposes.
  • Placing each redo log group on a separate disk that does not store any other files.
  • Striping the redo log files across several disks, using an operating system striping tool. (Manual striping is not possible in this situation.)
  • Avoiding the use of RAID 5 for redo logs.

Redo log files are written sequentially by the Log Writer (LGWR) process. This operation can be made faster if there is no concurrent activity on the same disk. Dedicating a separate disk to redo log files usually ensures that LGWR runs smoothly with no further tuning necessary. If your system supports asynchronous I/O but this feature is not currently configured, then test to see if using this feature is beneficial. Performance bottlenecks related to LGWR are rare.

http://docs.oracle.com/cd/B19306_01/server.102/b14211/iodesign.htm