Topic
3 replies Latest Post - ‏2012-08-14T21:19:54Z by svadlamu
Erwin_Karbasi
Erwin_Karbasi
186 Posts
ACCEPTED ANSWER

Pinned topic Is write behind can lower retrieving latency

‏2012-08-02T16:18:43Z |
Hello Masters,

We are using cache with optimistic lock strategy.
We have PBE (provisioning) applications that write entries into the cache.
We also have other telephony application that read data from the the cache.
The reader application should have pretty low latency and high performance.

The question is, whether using Write-behind method can improve/lower the latency of reader application (reading data from the cache) in any cases?

Thank you in advance,
Erwin
Updated on 2012-08-14T21:19:54Z at 2012-08-14T21:19:54Z by svadlamu
  • Erwin_Karbasi
    Erwin_Karbasi
    186 Posts
    ACCEPTED ANSWER

    Re: Is write behind can lower retrieving latency

    ‏2012-08-08T05:15:14Z  in response to Erwin_Karbasi
    Your insight would be highly appreciated.
    • Erwin_Karbasi
      Erwin_Karbasi
      186 Posts
      ACCEPTED ANSWER

      Re: Is write behind can lower retrieving latency

      ‏2012-08-14T13:53:32Z  in response to Erwin_Karbasi
      Please help us.
      • svadlamu
        svadlamu
        3 Posts
        ACCEPTED ANSWER

        Re: Is write behind can lower retrieving latency

        ‏2012-08-14T21:19:54Z  in response to Erwin_Karbasi
        Hello Erwin,

        I will give you this paragraph from WXS v8.5 InfoCenter, that might help provide a pointer:

        Write-behind caching performance considerations

        Write-behind caching support improves response time by removing the loader update from the transaction. It also increases database throughput because database updates are combined. It is important to understand the overhead introduced by write-behind thread, which pulls the data out of the queue map and pushed to the loader.

        The maximum update count or the maximum update time need to be adjusted based on the expected usage patterns and environment. If the value of the maximum update count or the maximum update time is too small, the overhead of the write-behind thread may exceed the benefits. Setting a large value for these two parameters could also increase the memory usage for queuing the data and increase the stale time of the database records.

        For best performance, tune the write-behind parameters based on the following factors:

        * Ratio of read and write transactions
        * Same record update frequency
        * Database update latency.
        http://pic.dhe.ibm.com/infocenter/wxsinfo/v8r5/topic/com.ibm.websphere.extremescale.doc/cxswritebehapp.html

        Regards,
        • Srini.

        WebSphere eXtreme Scale (WXS) Support.

        IBM
        RTP, NC 27709