lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAOzFzEjJSuTKy9-wk-qSzsmMHfB8h2ushRG6LQ0+92uDyB4sZg@mail.gmail.com>
Date:	Sun, 1 Apr 2012 13:51:05 +1000
From:	Joseph Glanville <joseph.glanville@...onvm.com.au>
To:	Jeffrey Merkey <jeffmerkey@...il.com>
Cc:	David Dillow <dave@...dillows.org>,
	linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [ANNOUNCE] Linux hits 20 Gb/s stream to disk levels over Infiniband

On 1 April 2012 05:23, Jeffrey Merkey <jeffmerkey@...il.com> wrote:
> On 3/31/12, David Dillow <dave@...dillows.org> wrote:
>> On Sat, 2012-03-31 at 11:28 -0600, Jeffrey Merkey wrote:
>>> Made it to 20 Gb/s network packet capture stream to disk performance
>>> levels over Infiniband fabrics on Linux.
>>
>> I'm assuming that the interesting part is the 20 Gbits/s capture? And at
>> minimum packet sizes, I'd hope. Linux has been able to stream more than
>> 5 GBytes/s (40 Gbits/s) to disk for several years now over SRP.
>
> Network Capture while being routed and regenerated and piped through
> apps at the same time from live network traffic.  Most of the other
> vendors tout 10 Gb/s stream to disk but only Solera really can do it.
> All of them except Niksun are Linux and I think they moved some of
> their stuff to it.  So hitting 20 Gb/s is a big deal right now.  With
> Intel's push in IB/Ethernet, this is where the next thing is going.
> 3D Torus stuff is totally cool.  You can actually store data in the
> network itself.  Talk about a challenge for Network security.

You can do Infiniband routing in software? This is indeed very interesting.
Pinged your contact page. :)

>
> The next class of malicious programs may be adapted to live within
> fabrics and require new methods to detect and remove them.
>
>>
>>>  Will be trying this on 3.0
>>> after I work through the OFED issues (long list).  Post the results
>>> from 3.0.  These are from 2.6.XX series kernels.
>>
>> You may find it easier to use a recent RedHat or SUSE with an upstream
>> kernel. I know that RHEL6 userland is compatible with the upstream
>> kernel's Infiniband, and OFED is moving its kernel work to a "backport
>> upstream" model for OFED 3.2.
>>
>
>
> I've got RH6 done, but OFED was down on the list and SLES was ahead
> of.it.  No rest for the weary.
>
> Jeff
>
>>> Linux f_cking ROCKS!!!
>>
>> Indeed it does.
>>
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Joseph.

-- 
Founder | Director | VP Research
Orion Virtualisation Solutions | www.orionvm.com.au | Phone: 1300 56
99 52 | Mobile: 0428 754 846
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ