cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ariel Weisberg (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-8714) row-cache: use preloaded jemalloc w/ Unsafe
Date Fri, 06 Feb 2015 21:45:35 GMT


Ariel Weisberg commented on CASSANDRA-8714:

Something doesn't sit right with me for the difference between JNA accessing jemalloc via
JNI and Unsafe. Since you are benchmarking access path I think you only need to allocate 1-byte.
Make sure to store a value to the location so the page is touched.

If Unsafe is slower I would be comfortable with using JNA to allocate.

You mention address translation (addr_to_java), but that is just a cast to get things to compile.
I followed it through to os.cpp and I don't see any real extra work.

It's quite possible the results are accurate and JNI just ended up better optimized then the
native function calls in Unsafe. JNI has received a lot of attention.

> row-cache: use preloaded jemalloc w/ Unsafe
> -------------------------------------------
>                 Key: CASSANDRA-8714
>                 URL:
>             Project: Cassandra
>          Issue Type: Sub-task
>            Reporter: Robert Stupp
>            Assignee: Robert Stupp
>             Fix For: 3.0
>         Attachments: 8714.txt
> Using jemalloc via Java's {{Unsafe}} is a better alternative on Linux than using jemalloc
via JNA.

This message was sent by Atlassian JIRA

View raw message