Open Stack

----- Original Message -----
> From: "Donald H Villeneuve" <donald.h.villeneuve at intel.com>
> To: openstack-hpc at lists.openstack.org
> 
> Hi,
> New to this list,  I would like to attend any HPC oriented meetings or
> presentation there may be in Vancouver.
> If to attend, some confirmation needs to be sent, please advise.
> 
> I've been tasked to look into the provision of cluster with large node count.
> So far many ideas to make it happen.
> But I'm still learning my way around OpenStack.  I'll be able to provide more
> details as a measure of success is achieved.
> 
> Regards,
> Donald Villeneuve

+1, I'm also interested in attending. I am not coming at it so much from a HPC operator point of view (though I did work in the area in a previous life) but from an interest in understanding how HPC operators are using OpenStack, and what further improvements they would like to see.

Thanks,

Steve

> -----Original Message-----
> From: openstack-hpc-request at lists.openstack.org
> [mailto:openstack-hpc-request at lists.openstack.org]
> Sent: Thursday, April 16, 2015 10:36 PM
> To: openstack-hpc at lists.openstack.org
> Subject: OpenStack-HPC Digest, Vol 23, Issue 4
> 
> Send OpenStack-HPC mailing list submissions to
> 	openstack-hpc at lists.openstack.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-hpc
> or, via email, send a message with subject or body 'help' to
> 	openstack-hpc-request at lists.openstack.org
> 
> You can reach the person managing the list at
> 	openstack-hpc-owner at lists.openstack.org
> 
> When replying, please edit your Subject line so it is more specific than "Re:
> Contents of OpenStack-HPC digest..."
> 
> 
> Today's Topics:
> 
>    1. Re: +1 for attendance to the HPC get-together	in	Vancouver.
>       (Tim Bell)
>    2. Re: +1 for attendance to the HPC get-together in	Vancouver.
>       (Matt Jarvis)
>    3. Re: +1 for attendance to the HPC get-together in	Vancouver.
>       (Matt Jarvis)
>    4. Re: CPU intensive apps on OpenStack (Blair Bethwaite)
> 
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Thu, 16 Apr 2015 13:02:17 +0000
> From: Tim Bell <Tim.Bell at cern.ch>
> To: Matt Jarvis <matt.jarvis at datacentred.co.uk>,
> 	"openstack-hpc at lists.openstack.org"
> 	<openstack-hpc at lists.openstack.org>
> Subject: Re: [openstack-hpc] +1 for attendance to the HPC get-together
> 	in	Vancouver.
> Message-ID:
> 	<5D7F9996EA547448BC6C54C8C5AAF4E501029CF555 at CERNXCHG43.cern.ch>
> Content-Type: text/plain; charset="utf-8"
> 
> Can you send round the slides URL ?
> 
> Thanks
> Tim
> 
> From: Matt Jarvis [mailto:matt.jarvis at datacentred.co.uk]
> Sent: 16 April 2015 12:28
> To: openstack-hpc at lists.openstack.org
> Subject: [openstack-hpc] +1 for attendance to the HPC get-together in
> Vancouver.
> 
> Hi, new to this list, but we're running ATLAS workloads for the University of
> Lancaster on our OpenStack public cloud in the UK, and actively looking to
> run more HPC/HTC type workloads. Would be interested in attending a session
> in Vancouver. Peter Love from Lancaster University recently presented at
> HEPIX in Oxford on some of the challenges around his work on our cloud, with
> comparison to grid sites. I'd be happy to share the slides if anyone is
> interested.
> 
> --
> Matt Jarvis
> Head of Cloud Computing
> DataCentred
> Office: (+44)0161 8703985
> Mobile: (+44)07983 725372
> Email: matt.jarvis at datacentred.co.uk<mailto:matt.jarvis at datacentred.co.uk>
> Website: http://www.datacentred.co.uk
> 
> DataCentred Limited registered in England and Wales no. 05611763
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://lists.openstack.org/pipermail/openstack-hpc/attachments/20150416/46124136/attachment-0001.html>
> 
> ------------------------------
> 
> Message: 2
> Date: Thu, 16 Apr 2015 14:05:59 +0100
> From: Matt Jarvis <matt.jarvis at datacentred.co.uk>
> To: Tim Bell <Tim.Bell at cern.ch>
> Cc: "openstack-hpc at lists.openstack.org"
> 	<openstack-hpc at lists.openstack.org>
> Subject: Re: [openstack-hpc] +1 for attendance to the HPC get-together
> 	in	Vancouver.
> Message-ID:
> 	<CACpXDz-5d3Mqztr_OuxNzdT4XiXHSgmB-2zajfQcj5yOUaeeJA at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> https://indico.cern.ch/event/346931/session/5/contribution/25
> 
> On 16 April 2015 at 14:02, Tim Bell <Tim.Bell at cern.ch> wrote:
> 
> >  Can you send round the slides URL ?
> >
> >
> >
> > Thanks
> >
> > Tim
> >
> >
> >
> > *From:* Matt Jarvis [mailto:matt.jarvis at datacentred.co.uk]
> > *Sent:* 16 April 2015 12:28
> > *To:* openstack-hpc at lists.openstack.org
> > *Subject:* [openstack-hpc] +1 for attendance to the HPC get-together
> > in Vancouver.
> >
> >
> >
> > Hi, new to this list, but we're running ATLAS workloads for the
> > University of Lancaster on our OpenStack public cloud in the UK, and
> > actively looking to run more HPC/HTC type workloads. Would be
> > interested in attending a session in Vancouver. Peter Love from
> > Lancaster University recently presented at HEPIX in Oxford on some of
> > the challenges around his work on our cloud, with comparison to grid
> > sites. I'd be happy to share the slides if anyone is interested.
> >
> >
> > --
> >
> > Matt Jarvis
> >
> > Head of Cloud Computing
> > DataCentred
> >
> > Office: (+44)0161 8703985
> > Mobile: (+44)07983 725372
> >
> > Email: matt.jarvis at datacentred.co.uk
> >
> > Website: http://www.datacentred.co.uk
> >
> >
> > DataCentred Limited registered in England and Wales no. 05611763
> >
> 
> 
> 
> --
> Matt Jarvis
> Head of Cloud Computing
> DataCentred
> Office: (+44)0161 8703985
> Mobile: (+44)07983 725372
> Email: matt.jarvis at datacentred.co.uk
> Website: http://www.datacentred.co.uk
> 
> --
> DataCentred Limited registered in England and Wales no. 05611763
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://lists.openstack.org/pipermail/openstack-hpc/attachments/20150416/bbc2ebe0/attachment-0001.html>
> 
> ------------------------------
> 
> Message: 3
> Date: Thu, 16 Apr 2015 20:26:46 +0100
> From: Matt Jarvis <matt.jarvis at datacentred.co.uk>
> To: Tim Bell <Tim.Bell at cern.ch>
> Cc: "openstack-hpc at lists.openstack.org"
> 	<openstack-hpc at lists.openstack.org>
> Subject: Re: [openstack-hpc] +1 for attendance to the HPC get-together
> 	in	Vancouver.
> Message-ID:
> 	<CACpXDz-NyVPx6FgcCDzk55jbrjmLYZx4YyKXfk8gO=VDKJNYfA at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Probably also worth pointing out that the cores limit is somewhat artificial,
> Lancaster in this phase have a quota limit of 200 cores available to them.
> 
> On 16 April 2015 at 14:05, Matt Jarvis <matt.jarvis at datacentred.co.uk>
> wrote:
> 
> > https://indico.cern.ch/event/346931/session/5/contribution/25
> >
> > On 16 April 2015 at 14:02, Tim Bell <Tim.Bell at cern.ch> wrote:
> >
> >>  Can you send round the slides URL ?
> >>
> >>
> >>
> >> Thanks
> >>
> >> Tim
> >>
> >>
> >>
> >> *From:* Matt Jarvis [mailto:matt.jarvis at datacentred.co.uk]
> >> *Sent:* 16 April 2015 12:28
> >> *To:* openstack-hpc at lists.openstack.org
> >> *Subject:* [openstack-hpc] +1 for attendance to the HPC get-together
> >> in Vancouver.
> >>
> >>
> >>
> >> Hi, new to this list, but we're running ATLAS workloads for the
> >> University of Lancaster on our OpenStack public cloud in the UK, and
> >> actively looking to run more HPC/HTC type workloads. Would be
> >> interested in attending a session in Vancouver. Peter Love from
> >> Lancaster University recently presented at HEPIX in Oxford on some of
> >> the challenges around his work on our cloud, with comparison to grid
> >> sites. I'd be happy to share the slides if anyone is interested.
> >>
> >>
> >> --
> >>
> >> Matt Jarvis
> >>
> >> Head of Cloud Computing
> >> DataCentred
> >>
> >> Office: (+44)0161 8703985
> >> Mobile: (+44)07983 725372
> >>
> >> Email: matt.jarvis at datacentred.co.uk
> >>
> >> Website: http://www.datacentred.co.uk
> >>
> >>
> >> DataCentred Limited registered in England and Wales no. 05611763
> >>
> >
> >
> >
> > --
> > Matt Jarvis
> > Head of Cloud Computing
> > DataCentred
> > Office: (+44)0161 8703985
> > Mobile: (+44)07983 725372
> > Email: matt.jarvis at datacentred.co.uk
> > Website: http://www.datacentred.co.uk
> >
> 
> 
> 
> --
> Matt Jarvis
> Head of Cloud Computing
> DataCentred
> Office: (+44)0161 8703985
> Mobile: (+44)07983 725372
> Email: matt.jarvis at datacentred.co.uk
> Website: http://www.datacentred.co.uk
> 
> --
> DataCentred Limited registered in England and Wales no. 05611763
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <http://lists.openstack.org/pipermail/openstack-hpc/attachments/20150416/f4aeb1b7/attachment-0001.html>
> 
> ------------------------------
> 
> Message: 4
> Date: Fri, 17 Apr 2015 15:36:02 +1000
> From: Blair Bethwaite <blair.bethwaite at gmail.com>
> To: openstack-hpc at lists.openstack.org, Tim Bell <Tim.Bell at cern.ch>
> Subject: Re: [openstack-hpc] CPU intensive apps on OpenStack
> Message-ID:
> 	<CA+z5DsxMJkD1K6uieiHCLprodiT35nDs6hW8C3Tg9=OW-KLUrA at mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
> 
> Hi Tim,
> 
> What's your reference on disabling EPT? I can find some fairly old stuff on
> this, but I thought newer processors had improved performance here to lower
> page fault overheads... so I guess I'm wondering if you've seen a recent
> evaluation somewhere?
> 
> Re. NUMA etc, it seems like the most important thing if you're running large
> memory apps inside your guests is to make sure the guest is pinned (both CPU
> and NUMA wise) and sees a NUMA topology that matches how it is pinned on the
> host. I haven't had a chance to try it yet but I thought this was all now
> possible Juno. Beyond that you probably want the guest kernel to be NUMA
> sassy as well, so it needs (IIRC) Linux 3.13+ to get the numa_balancing (nee
> autonuma) goodies. I'm not sure if for such cases you might actually want to
> disable numa balancing on the host or whether the pinning would effectively
> remove any overhead there anyway...
> 
> PS: +1 on the meetup suggestion
> 
> --
> Cheers,
> ~Blairo
> 
> 
> 
> ------------------------------
> 
> _______________________________________________
> OpenStack-HPC mailing list
> OpenStack-HPC at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-hpc
> 
> 
> End of OpenStack-HPC Digest, Vol 23, Issue 4
> ********************************************
> 
> _______________________________________________
> OpenStack-HPC mailing list
> OpenStack-HPC at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-hpc
> 

-- 
Steve Gordon, RHCE
Sr. Technical Product Manager,
Red Hat Enterprise Linux OpenStack Platform