[jira] [Updated] (CASSANDRA-7438) Serializing Row cache alternative (Fully off heap)

["Vijay (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-7438?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vijay updated CASSANDRA-7438:
-----------------------------
    Attachment: tests.zip

Hi Jonathan, We should at least make the cache more pluggable with this ticket.

{quote}Well, it can, but it's almost always a bad idea. Not something we should optimize for.{quote}
Just my 2 cents: I have seen some use cases which store bigger blobs as column values too...

Sorry to diverge the discussion again. Alright, completed most of the items in this ticket. 
* Expiry thread to proactively remove items (looks like it catches up withe load pretty well).
* Some minor comments on metrics and a bit of refactor. (will have to revisit all the discussion again)
* Some error handling and additional tests.

Looks like lruc JNI based and Unsafe are close to each other. Numbers attached...
* The difference is that unsafe is performing worse at P99 distributions and C implementation at P999.
* I think the performance hit on both caches are during rehash global locks.
* Note these numbers are in my laptop and during the run's CPU's are 100% so might not be perfect.
* Will have to run benchmark on server grade (big memory) machines on Monday.

> Serializing Row cache alternative (Fully off heap)
> --------------------------------------------------
>
>                 Key: CASSANDRA-7438
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7438
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>         Environment: Linux
>            Reporter: Vijay
>            Assignee: Vijay
>              Labels: performance
>             Fix For: 3.0
>
>         Attachments: 0001-CASSANDRA-7438.patch, tests.zip
>
>
> Currently SerializingCache is partially off heap, keys are still stored in JVM heap as BB, 
> * There is a higher GC costs for a reasonably big cache.
> * Some users have used the row cache efficiently in production for better results, but this requires careful tunning.
> * Overhead in Memory for the cache entries are relatively high.
> So the proposal for this ticket is to move the LRU cache logic completely off heap and use JNI to interact with cache. We might want to ensure that the new implementation match the existing API's (ICache), and the implementation needs to have safe memory access, low overhead in memory and less memcpy's (As much as possible).
> We might also want to make this cache configurable.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)