Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-24 Thread Sylvain Bauza
Le 23/01/2014 18:17, Day, Phil a écrit : Just to be clear I'm not advocating putting any form of automated instance life-cycle into Nova - I agree that belongs in an external system like Climate. However for a reservations model to work efficiently it seems to be you need two complementary ty

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-23 Thread Jay Pipes
On Thu, 2014-01-23 at 17:05 +, Day, Phil wrote: > > -Original Message- > > From: Jay Pipes [mailto:jaypi...@gmail.com] > > Sent: 22 January 2014 02:01 > > To: openstack-dev@lists.openstack.org > > Subject: Re: [openstack-dev] Next steps for Whole Host al

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-23 Thread Day, Phil
Jay Pipes wrote on 01/21/2014 08:50:36 PM: ... > On Tue, 2014-01-21 at 14:28 +, Day, Phil wrote: > > > > > > > > I think there is clear water between this and the existing aggregate > > > > based > > > isolation.  I also think this is a different use case from reservations.   > > > It's > >

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-23 Thread Day, Phil
> -Original Message- > From: Sylvain Bauza [mailto:sylvain.ba...@bull.net] > Sent: 22 January 2014 10:24 > To: OpenStack Development Mailing List (not for usage questions) > Subject: Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds > > Le 22/01/20

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-23 Thread Sylvain Bauza
Le 23/01/2014 18:05, Day, Phil a écrit : I think there is a cross-over point here somewhere - call it "managed cloud" where the user doesn't want to run a cloud but does want (and is willing to pay for) some guarantee of isolation of hypervisors, but also wants to share other aspects of the clo

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-23 Thread Day, Phil
> -Original Message- > From: Jay Pipes [mailto:jaypi...@gmail.com] > Sent: 22 January 2014 02:01 > To: openstack-dev@lists.openstack.org > Subject: Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds > > On Tue, 2014-01-21 at 14:21 +,

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-22 Thread Sylvain Bauza
Le 22/01/2014 15:57, Mike Spreitzer a écrit : Is that enough? Remember that some of us are concerned with business workloads, rather than HPC jobs. While it might be acceptable in a business workload to plan on regularly recycling every individual instance, it is definitely not acceptable t

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-22 Thread Mike Spreitzer
Jay Pipes wrote on 01/21/2014 08:50:36 PM: ... > On Tue, 2014-01-21 at 14:28 +, Day, Phil wrote: > > > > > > > > I think there is clear water between this and the existing > aggregate based > > > isolation. I also think this is a different use case from > reservations. It's > > > *mostly*

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-22 Thread Mike Spreitzer
> From: Sylvain Bauza > > Le 22/01/2014 02:50, Jay Pipes a écrit : > > > > Yup, agreed. It's difficult to guess what the capacity implications > > would be without having solid numbers on customer demands for this > > functionality, including hard data on how long such instances would > > typical

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-22 Thread Sylvain Bauza
Le 22/01/2014 02:50, Jay Pipes a écrit : Yup, agreed. It's difficult to guess what the capacity implications would be without having solid numbers on customer demands for this functionality, including hard data on how long such instances would typically live (see my previous point about re-using

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Jay Pipes
On Tue, 2014-01-21 at 14:21 +, Khanh-Toan Tran wrote: > > Exactly - that's why I wanted to start this debate about the way forward > > for the > > Pcloud Blueprint, which was heading into some kind of middle ground. As > > per > > my original post, and it sounds like the three of us are at l

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Jay Pipes
On Tue, 2014-01-21 at 14:28 +, Day, Phil wrote: > > > > > > I think there is clear water between this and the existing aggregate based > > isolation. I also think this is a different use case from reservations. > > It's > > *mostly* like a new scheduler hint, but because it has billing impa

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Jay Pipes
On Tue, 2014-01-21 at 11:57 +, Day, Phil wrote: > > > > So, I actually don't think the two concepts (reservations and > > > > "isolated instances") are competing ideas. Isolated instances are > > > > actually not reserved. They are simply instances that have a > > > > condition placed on their

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Day, Phil
> -Original Message- > From: Khanh-Toan Tran [mailto:khanh-toan.t...@cloudwatt.com] > Sent: 21 January 2014 14:21 > To: OpenStack Development Mailing List (not for usage questions) > Subject: Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds > > >

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Sylvain Bauza
Le 21/01/2014 15:28, Day, Phil a écrit : I think there is clear water between this and the existing aggregate based isolation. I also think this is a different use case from reservations. It's *mostly* like a new scheduler hint, but because it has billing impacts I think it needs to be more

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Mike Spreitzer
I want to explore the linkage between the proposed exclusive use filter and scheduling a little more. At the Icehouse summit I heard the concern that without advance preparation, there may well be zero hosts that are acceptable. Why is that not a concern now? What does the Nova scheduler do n

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Day, Phil
> > > > I think there is clear water between this and the existing aggregate based > isolation. I also think this is a different use case from reservations. > It's > *mostly* like a new scheduler hint, but because it has billing impacts I > think it > needs to be more than just that - for exa

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Khanh-Toan Tran
> Exactly - that's why I wanted to start this debate about the way forward > for the > Pcloud Blueprint, which was heading into some kind of middle ground. As > per > my original post, and it sounds like the three of us are at least aligned > I'm > proposing to spilt this into two streams: > >

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Sylvain Bauza
Le 21/01/2014 12:57, Day, Phil a écrit : So, I actually don't think the two concepts (reservations and "isolated instances") are competing ideas. Isolated instances are actually not reserved. They are simply instances that have a condition placed on their assignment to a particular compute node t

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-21 Thread Day, Phil
> > > So, I actually don't think the two concepts (reservations and > > > "isolated instances") are competing ideas. Isolated instances are > > > actually not reserved. They are simply instances that have a > > > condition placed on their assignment to a particular compute node > > > that the node

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Sylvain Bauza
2014/1/20 Jay Pipes > > > But I believe that the two concerns can be tackled separately. > > Indeed. I fully agree with the fact isolation can be provided by Nova, and Climate would by the way happy to leverage it for providing capacity planning on top of it. By the way, I'm personnally convicted

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Jay Pipes
On Mon, 2014-01-20 at 17:49 +0100, Sylvain Bauza wrote: > Hi Jay, > > Le 20/01/2014 17:34, Jay Pipes a écrit : > > > On Mon, Jan 20, 2014 at 11:18 AM, Sylvain Bauza > > wrote: > > Jay, please be aware of the existence of Climate, which is a > > Stackforge project for managing ded

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Sylvain Bauza
Hi Jay, Le 20/01/2014 17:34, Jay Pipes a écrit : On Mon, Jan 20, 2014 at 11:18 AM, Sylvain Bauza mailto:sylvain.ba...@bull.net>> wrote: Jay, please be aware of the existence of Climate, which is a Stackforge project for managing dedicated resources (like AWS reserved instances). Th

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Jay Pipes
On Mon, Jan 20, 2014 at 11:18 AM, Sylvain Bauza wrote: > Jay, please be aware of the existence of Climate, which is a Stackforge > project for managing dedicated resources (like AWS reserved instances). > This is not another API extension, but another API endpoint for creating > what we call "leas

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Sylvain Bauza
Le 20/01/2014 16:57, Jay Pipes a écrit : On Mon, Jan 20, 2014 at 10:18 AM, Day, Phil > wrote: HI Folks, The original (and fairly simple) driver behind whole-host-allocation (https://wiki.openstack.org/wiki/WholeHostAllocation) was to enable users to

Re: [openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Jay Pipes
On Mon, Jan 20, 2014 at 10:18 AM, Day, Phil wrote: > HI Folks, > > > > The original (and fairly simple) driver behind whole-host-allocation ( > https://wiki.openstack.org/wiki/WholeHostAllocation) was to enable users > to get guaranteed isolation for their instances. This then grew somewhat > a

[openstack-dev] Next steps for Whole Host allocation / Pclouds

2014-01-20 Thread Day, Phil
HI Folks, The original (and fairly simple) driver behind whole-host-allocation (https://wiki.openstack.org/wiki/WholeHostAllocation) was to enable users to get guaranteed isolation for their instances. This then grew somewhat along the lines of "If they have in effect a dedicated hosts then wo