Re: MySQL Scalability issues due to glibc

Hi,

YFYI:

It seems that the famous MySQL scalability issues on SMP is due to a
glibc deficiency:

http://jeffr-tech.livejournal.com/6268.html#cutid1
http://ozlabs.org/~anton/linux/sysbench/

So we should stop beating on MySQL for this issue. :-)

HTH,
Markus
--
Markus Schaber | Logical Tracking&Tracing International AG
Dipl. Inf. | Software Development GIS

Fight against software patents in Europe! www.ffii.org
www.nosoftwarepatents.org

Markus Schaber wrote:
> Hi,
>
> YFYI:
>
> It seems that the famous MySQL scalability issues on SMP is due to a
> glibc deficiency:
>
> http://jeffr-tech.livejournal.com/6268.html#cutid1
> http://ozlabs.org/~anton/linux/sysbench/
>
> So we should stop beating on MySQL for this issue. :-)

That's ok - there are plenty more :-)

/D

Probably should be posting this to www, but:

When I go to Pgfoundry, the page is in Spanish. Traceroute shows it
bouncing around in Panama a bit.
After 55 years, I'm just beginning to get a reasonable grasp of English.

Happy St. Patrick's Day Eve,
Mike Ellsworth

Markus Schaber wrote:
> Hi,
>
> YFYI:
>
> It seems that the famous MySQL scalability issues on SMP is due to a
> glibc deficiency:
>
> http://jeffr-tech.livejournal.com/6268.html#cutid1
> http://ozlabs.org/~anton/linux/sysbench/
>
> So we should stop beating on MySQL for this issue. :-)

Does that mean we can beat on GNU now?

>
>
> HTH,
> Markus

--

=== The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564 || 24x7/Emergency: +1.800.492.2240
Providing the most comprehensive PostgreSQL solutions since 1997
http://www.commandprompt.com/

Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
PostgreSQL Replication: http://www.commandprompt.com/products/

nhrcommu(at)rochester(dot)rr(dot)com wrote:
> Probably should be posting this to www, but:
>
> When I go to Pgfoundry, the page is in Spanish. Traceroute shows it
> bouncing around in Panama a bit.
> After 55 years, I'm just beginning to get a reasonable grasp of English.
>

This appears to have cleaned itself up?

> Happy St. Patrick's Day Eve,
> Mike Ellsworth
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Have you searched our list archives?
>
> http://archives.postgresql.org
>

--

=== The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564 || 24x7/Emergency: +1.800.492.2240
Providing the most comprehensive PostgreSQL solutions since 1997
http://www.commandprompt.com/

Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
PostgreSQL Replication: http://www.commandprompt.com/products/

> > When I go to Pgfoundry, the page is in Spanish. Traceroute shows it
> > bouncing around in Panama a bit.
> > After 55 years, I'm just beginning to get a reasonable grasp of
> English.>
>
> This appears to have cleaned itself up?
>

Yup.

Thanks

Hi, Joshua,

"Joshua D. Drake" <jd(at)commandprompt(dot)com> wrote:

> > It seems that the famous MySQL scalability issues on SMP is due to a
> > glibc deficiency:
> >
> > http://jeffr-tech.livejournal.com/6268.html#cutid1
> > http://ozlabs.org/~anton/linux/sysbench/
> >
> > So we should stop beating on MySQL for this issue. :-)
>
> Does that mean we can beat on GNU now?

Feel free to do that, and ask People like Fefe (DietLibC author) for
permission. :-)

Regards,
Markus

--
Markus Schaber | Logical Tracking&Tracing International AG
Dipl. Inf. | Software Development GIS

Fight against software patents in Europe! www.ffii.org
www.nosoftwarepatents.org

All,

>>> It seems that the famous MySQL scalability issues on SMP is due to a
>>> glibc deficiency:
>>>
>>> http://jeffr-tech.livejournal.com/6268.html#cutid1
>>> http://ozlabs.org/~anton/linux/sysbench/

Note that sysbench is specifically a no-db-lock-contention benchmark.
MySQL+InnoDB does still have a lock contention scalability issue (as do
we, at a higher # of processors).

--Josh

On 3/21/07, Josh Berkus <josh(at)agliodbs(dot)com> wrote:
> Note that sysbench is specifically a no-db-lock-contention benchmark.
> MySQL+InnoDB does still have a lock contention scalability issue (as do
> we, at a higher # of processors).

True, but no one seems to do general benchmarks with InnoDB outside of
Linux. Have you guys at Sun run them? If so, what performance do you
see with MySQL+InnoDB on Solaris?

--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
33 Wood Ave S, 3rd Floor | jharris(at)enterprisedb(dot)com
Iselin, New Jersey 08830 | http://www.enterprisedb.com/

Jonah,

> True, but no one seems to do general benchmarks with InnoDB outside of
> Linux. Have you guys at Sun run them? If so, what performance do you
> see with MySQL+InnoDB on Solaris?
>

Scaling stops and begins to decline after 8 cores on Appserver
workloads. For PostgreSQL, it's closer to 32.

--Josh Berkus