High Utilization and IOwait% on Clariion Thin LUN

Hi All,

Storage Box:- Clariion CX4-240

FLARE OE:- 4.30

Total memory in each SP:- 4064 MB

Free Memory	0 MB	0 MB
Read Cache Size	253 MB	253 MB
Write Cache Size	1008 MB	1008 MB
Cache Page Size	8 KB	8 KB
Low Watermark	60%	60%
High Watermark	80%	80%
HA Cache Vault	Disabled	Disabled

Issue summary:- Suddenly for one of the SAP customer's application IOwait% increases above 90% most of the time. LUN utilization from storage perspective also almost above 50% most of the time. IOwait% from SAP server's LINUX O.S nearby 67% - 75%.

Step1:- We had checked with REDHAT,  as per them below comments they have given :-   “It can be seen from the above collected data that read from the device or write to the device are not very high. However await is increasing which should not be the case for such nominal values. Now as we know Await basically tells the average time (in milliseconds)  for I/O requests issued to the device to be served. This includes the time  spent by the requests in queue and the time spent servicing them. That is whatever await is being seen is due to the time taken by the storage to complete the I/O.”

Step2:- We had migrated Old Thin LUN to new Thin LUN but no luck. We tried to create thick LUN but it was trespassing. From EMC team prespective, as per the SP collects analysis, there are no hardware faults or critical errors found.

Don't know where is the problem. .  Kindly advice me what to do. Need help.

Question:- 1. We are using cache mirroring then Read cache size and write cache size reduced due to cache mirroring. Is there any possibility, to disable the cache mirroring and enable cache vault then we will get full 4GB for read (2GB) and write (2GB) cache. ??