[Oisf-devel] GPU Support
Charles Wyble
charles at thewybles.com
Wed Jan 6 20:07:22 UTC 2010
Ah nice. There is an opencl implementation for cell:
http://alphaworks.ibm.com/tech/opencl
So yes I have ATI/ps3/Nvidia hardware and will be happy to test and
report. :)
On 1/6/10 7:50 AM, Victor Julien wrote:
> Hi Charles,
>
> Charles Wyble wrote:
>> Hello,
>>
>> My name is Charles Wyble and I'm a systems/network engineer. I've long
>> been interested in security as a practical matter. Worked for numerous
>> e-commerce and other very sensitive environments.
>>
>> I'm currently in the process of setting up a very comprehensive lab
>> environment and plan to use Suricata IDS for the IDS/IPS. Thus far I've
>> just been researching it, and it looks very promising. Plan to combine
>> it with opsview (already in use) , to get a comprehensive overview of
>> the network and show that open source net security/monitoring tools are
>> awesome.
>>
>> Anyway, I noticed in the release announcement discussion of GPU
>> offloading. This is a huge passion of mine, and I would like to help.
>>
>> I have a CUDA Nvidia GPU, (01:00.0 VGA compatible controller: nVidia
>> Corporation G98 [GeForce 8400 GS] (rev a1)
>> a ps3 running Ubuntu 9.04 (cell sdk) and an ATI stream card (brook)
>>
>> I'm fairly new to CUDA/OpenCL development, but would be happy to
>> alpha/beta test code. Or if developers need access to hardware for
>> testing purposes I would be happy to set that up.
>
> Your help will be greatly appreciated!
>
> The current status of our support is that we have a almost working
> OpenCL pattern matcher. Almost, as we keep running into driver issues
> and such. The status of OpenCL implementations so far seems a little
> immature.
>
> This is why we plan to focus on CUDA first the coming weeks. That will
> limit us to Nvidia obviously, but at least it should work then :)
>
> A second issue is that just invoking the GPU based pattern matcher for
> each packet/buffer is going to kill performance, not help it. So we'll
> have to design some sort of batch pattern scan mode. That too is
> something I plan to work on in the coming weeks.
>
> So... lot of work to do :)
>
> I will commit our current OpenCL soon which means you can look at it and
> run it to see if it has bugs (assuming we get rid of the current bugs,
> see above) and then complain about the performance :) Later I hope our
> code will be more useful and then any testing will be greatly appreciated!
>
> Thanks,
> Victor
>
More information about the Oisf-devel
mailing list