ULE vs. 4BSD scheduler benchmarks

O. Hartmann ohartman at zedat.fu-berlin.de
Sun Jan 29 10:12:40 UTC 2012

Am 01/28/12 23:40, schrieb Florian Smeets:
> [current@ bcc'ed to get a wider audience, please discuss on performance@]
> Hi,
> in recent times i saw a lot of threads where it was suggested people
> should switch from the ULE to the 4BSD scheduler. That got me thinking
> and i decided to run a few benchmarks. I looked through all the stuff
> Kris and Jeff did a few years ago and tried to follow their example. The
> main motivation is however that we (Attilio Rao and I) want to set a
> baseline for future reference, mainly for the work that's going on in
> the vmcontention branch right now, that is the reason why all tests were
> run on head at r229659. All debugging was disabled (WITNESS and friends for
> the kernel and MALLOC_PRODUCTION=yes for libc).
> For now i ran 3 different things. MySQL/sysbench, PostgreSQL/pgbench and
> pbzip2.
> All software was installed from ports with the default system gcc (gcc
> version 4.2.1 20070831 patched [FreeBSD]), with the exception of
> PostgreSQL. I created new postgres92-{server,client} ports with a
> snapshot of PostgreSQL 9.2dev from 16.01.2012, as a lot of scalability
> work was done in PostgreSQL 9.2.
> MySQL version 5.5.20
> sysbench version 0.4.12
> PostgreSQL/pgbench version 9.2dev
> PBZIP2 version v1.1.6
> The machine these test were run on is a 2x4 core Xeon L5310  @ 1.60GHz
> with 4GB RAM. Here is the complete topology:
> kern.sched.topology_spec: <groups>
>  <group level="1" cache-level="0">
>   <cpu count="8" mask="ff">0, 1, 2, 3, 4, 5, 6, 7</cpu>
>   <children>
>    <group level="2" cache-level="2">
>     <cpu count="4" mask="f">0, 1, 2, 3</cpu>
>    </group>
>    <group level="2" cache-level="2">
>     <cpu count="4" mask="f0">4, 5, 6, 7</cpu>
>    </group>
>   </children>
>  </group>
> </groups>
> The database benchmarks were all run with a work set that fit into the
> configured database memory, so after the warmup phase no disk io was
> involved. sysbench was run with 1 million rows, innodb was the engine we
> used as Kris work already showed that it scales much better than myisam
> (also innodb is the default in MySQL's 5.5 branch). Pgbench was run
> using a scaling factor of 100. The connection to the databases was using
> a unix socket, also only read only tests were run.
> The input and output files for the pbzip2 test were on tmpfs.
> The results are available in this Google docs spreadsheet, if you scroll
> down there are also some nice graphs.
> https://docs.google.com/spreadsheet/ccc?key=0Ai0N1xDe3uNAdDRxcVFiYjNMSnJWOTZhUWVWWlBlemc
> Over time i will add more benchmarks to the doc (i.e nginx/php-fpm and
> so on). I tried to run some nginx benchmarks, but those are limited by
> netisr, as i did not find a web server benchmark tool which can use unix
> sockets, any suggestions welcome.
> The conclusion right now seems to be that ULE is faster for database
> workload, but for strongly CPU-bound workloads 4BSD can be a better
> choice. I can provide KTR traces and/or schedgraph output for cases
> where 4BSD is better than ULE.
> I want to thank Sean Bruno and Yahoo for setting up / providing the
> machines to run these test on, and Attilio for suggestions and his
> general helpfulness.
> Florian

thanks a lot for your efforts. Nice work.

We got a new workstation, two socket 6-core westmere XEON's, I forgot
the specifications, but they're driven with 2,66 GHz each and have
access to 96GB RAM. Maybe I can also setup some benchmarks, but I need
advice since I'm not a kernel GURU.
The box is prmarily running Linux due to the TESLA/GPGPU stuff we run on
it. A colleague of mine developend a software for huge satellite imagery
correction needed in planetary science, the software is highly scalable
(OpenMP) and massively using OpenCL, but using OpenCL could be switched
off. We are not interested in database performance, but more in HPC
stuff and scientific calculations. I guess we could provode also some
benchmark results after a proper setup for the workload. Since this box
in question is also running a Linux Ubuntu 11.04 server, I would be
interesting having a comparison to that.

I intended to use FreeBSD 9.0-STABLE for benchmarking, but as I can see,
it is more applicable to use FreeBSD 10 since development goes on on
that branch.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
Url : http://lists.freebsd.org/pipermail/freebsd-performance/attachments/20120129/2c1bc7af/signature.pgp

More information about the freebsd-performance mailing list