sysbench-developers team mailing list archive

[Hakan Küçükyılmaz <hakan@xxxxxxxxxxxx>]

Alexey,

On 28.10.2010, at 21:07, Alexey Kopytov wrote:

> Review: Needs Information
> Hi Hakan,
> 
> Generally the patch looks good to me, thank you for the contribution.
> 
> I have only one question: why is thread_id rather than 1 passed as the lower endpoint to sb_rand() in both tests?

Sorry for confusion. I wanted to add some internal details about MariaDB's
new key_cache_segments feature for MyISAM, so that the distribution of
the selected ranges in the new tests make sense.

Full documentation of key_cache_segments:
   http://kb.askmonty.org/v/segmented-key-cache

What you need to know for the new sysbench tests:
"When working with a segmented key cache any key cache interface function that needs only one page has to acquire the key cache lock only for the segment the page is assigned to. This makes the chances for threads not having to compete for the same key cache lock better."

That means that the ranges should be different for every thread. I tried to
simulate that by
    params[i] = sb_rand(thread_id, oltp_table_size)

Maybe there is a better way to handle this?

Best,

Hakan

--
Hakan Küçükyılmaz, QA/Benchmark Engineer, Stuttgart/Germany
Monty Program Ab, http://askmonty.org/
Skype: hakank_  Phone: +49 171 1919839

-- 
https://code.launchpad.net/~hakan-askmonty/sysbench/hakan/+merge/39548
Your team sysbench-developers is subscribed to branch lp:sysbench.