| *** crushil_ has joined #openstack-cyborg | 00:11 | |
| *** crushil_ has quit IRC | 00:30 | |
| *** crushil_ has joined #openstack-cyborg | 00:31 | |
| *** crushil_ has quit IRC | 01:08 | |
| *** crushil_ has joined #openstack-cyborg | 03:14 | |
| *** crushil_ has quit IRC | 03:18 | |
| *** crushil_ has joined #openstack-cyborg | 03:21 | |
| *** crushil_ has quit IRC | 03:21 | |
| *** crushil_ has joined #openstack-cyborg | 03:25 | |
| *** crushil_ has quit IRC | 03:41 | |
| *** crushil_ has joined #openstack-cyborg | 03:55 | |
| *** crushil_ has quit IRC | 04:04 | |
| *** crushil_ has joined #openstack-cyborg | 04:04 | |
| *** crushil_ has quit IRC | 05:41 | |
| *** crushil_ has joined #openstack-cyborg | 05:43 | |
| *** crushil_ has quit IRC | 06:11 | |
| *** crushil_ has joined #openstack-cyborg | 08:43 | |
| *** crushil_ has quit IRC | 09:05 | |
| *** jkilpatr has quit IRC | 10:00 | |
| *** jkilpatr has joined #openstack-cyborg | 10:58 | |
| *** jkilpatr has quit IRC | 11:41 | |
| *** mikeH has joined #openstack-cyborg | 11:55 | |
| *** jkilpatr has joined #openstack-cyborg | 12:01 | |
| *** zhuli has quit IRC | 12:09 | |
| *** jkilpatr has quit IRC | 12:11 | |
| *** jkilpatr has joined #openstack-cyborg | 12:13 | |
| *** jkilpatr has quit IRC | 12:28 | |
| *** zhipeng has joined #openstack-cyborg | 14:02 | |
| *** joseppc has joined #openstack-cyborg | 14:03 | |
| *** joseppc has joined #openstack-cyborg | 14:03 | |
| *** zhipeng has quit IRC | 14:31 | |
| ttk2[m] | Morning everyone. | 14:59 |
|---|---|---|
| *** zhipeng has joined #openstack-cyborg | 15:00 | |
| zhipeng | hey | 15:02 |
| zhipeng | anyone else is here ? | 15:05 |
| *** NokMikeR has joined #openstack-cyborg | 15:06 | |
| zhipeng | #startmeeting openstack-cyborg | 15:13 |
| openstack | Meeting started Wed Jun 28 15:13:59 2017 UTC and is due to finish in 60 minutes. The chair is zhipeng. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:13 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:14 |
| openstack | The meeting name has been set to 'openstack_cyborg' | 15:14 |
| zhipeng | #topic Roll Call | 15:14 |
| NokMikeR | #info Michael Rooke, Nokia | 15:14 |
| zhipeng | #info Howard Huang | 15:15 |
| NokMikeR | Must be summer :) | 15:16 |
| ttk2[m] | #info Justin Kilpatrick, RedHat | 15:16 |
| ttk2[m] | Sorry on a plane today. | 15:16 |
| zhipeng | wow, still wifi ? | 15:16 |
| ttk2[m] | Just landed. | 15:17 |
| zhipeng | you work too hard :) | 15:17 |
| zhipeng | #topic development progress | 15:18 |
| ttk2[m] | Anyways I'd like to get that accelerator object stub merged so that I can start setting up the db components of the conductor. Messaging and rpc is working. | 15:18 |
| ttk2[m] | The agent needs the dummy driver merged at least so that I can play around with integrating those two. | 15:18 |
| ttk2[m] | Oh also I figured out the config module it loads transport urls and works in a real cloud env. | 15:19 |
| zhipeng | crushil how's the driver patch going ? | 15:19 |
| zhipeng | ttk2[m] config module ? | 15:19 |
| ttk2[m] | Oslo config. | 15:20 |
| zhipeng | okey | 15:20 |
| *** zhipeng has quit IRC | 15:21 | |
| ttk2[m] | Essentially I have all the basics worked out except the db so all the various components can talk to each other after that it's just a matter of writing business logic. | 15:21 |
| *** zhipeng has joined #openstack-cyborg | 15:21 | |
| *** zhipeng has quit IRC | 15:24 | |
| *** zhipeng has joined #openstack-cyborg | 15:24 | |
| zhipeng | ttk2[m] so looking at the conductor patch | 15:28 |
| zhipeng | #link https://review.openstack.org/#/c/472662/ | 15:28 |
| zhipeng | I think except for a need to modify the patch title, it looks good for me to merge | 15:29 |
| zhipeng | also the api spec could be merged at this point I guess | 15:30 |
| zhipeng | #link https://review.openstack.org/445814 | 15:30 |
| *** zhipeng has quit IRC | 15:31 | |
| *** zhipeng has joined #openstack-cyborg | 15:31 | |
| ttk2[m] | API spec is good. And I'll get the db stuff into the conductor and then mark it as ready. | 15:34 |
| zhipeng | cool | 15:34 |
| NokMikeR | does the api cover listing an application running on the accelerator? | 15:35 |
| zhipeng | we list accelerators, but i doubt we will list applications | 15:36 |
| zhipeng | could you give a use case on that ? | 15:36 |
| NokMikeR | GPU has a running application, you list the GPU to see if its 100% free, then discover its not, hence skip to the next one. | 15:36 |
| *** jkilpatr has joined #openstack-cyborg | 15:37 | |
| NokMikeR | thats more fine grained than simply discovering / listing the accelerator itself. | 15:37 |
| ttk2[m] | isn't that the same as listing usage? | 15:37 |
| ttk2[m] | I'm planning on usage metric collection but I'm not sure if/how we expose that to the api instead of just the operator/internally | 15:37 |
| zhipeng | i think we could definitely have that metric for the scheduler to use, but it might be just used internally | 15:38 |
| zhipeng | and it would be difficult to model it in the resource provider, because it is dynamic | 15:38 |
| NokMikeR | so operations like deleting an accelerator only occur when there is no running applications (=anything left) on the accelerator? | 15:39 |
| zhipeng | i think it would be up to the users to make that judgement (check if nothing left running) | 15:40 |
| ttk2[m] | that's easy enough to do as we have to control scheduling based on usage anyways | 15:41 |
| zhipeng | cyborg will just receive a request to detach the accelerator for whatever purpose | 15:41 |
| NokMikeR | ok | 15:41 |
| ttk2[m] | on the ohter hand it would be a pretty bad user experience if an idling VM kept the accelerator from being changed. | 15:41 |
| NokMikeR | or updated with something else since theres a left over app using resources that should be allocated for the new app etc. | 15:42 |
| NokMikeR | maybe these are issues outside of cyborg, not sure. | 15:42 |
| ttk2[m] | so managing instances attached to the accelerators is only partly cyborgs job | 15:51 |
| ttk2[m] | I mean we'll list instances that are attached or running on various accelerators, but deciding if the work being done is useful is out of scope. And in the case of shared accelerators we're going to need to rely on them being able to print out who is doing what since there's probably a scheduler running below us for that. | 15:52 |
| zhipeng | a sched running below us ? | 15:55 |
| ttk2[m] | zhipeng: like if we had gpu virtualization and three vm's sharing a gpu | 15:58 |
| ttk2[m] | there's a scheduler at the driver level taking workloads from the 3 virutal gpu's and muxing them into the real hardware. We may be able to see how much the actual hardware is utilized, but unless the driver tells us which vm is using what percentage of the total resources we're in the dark other than "these vm's are hooked up to this physical gpu" | 15:59 |
| zhipeng | agree | 16:06 |
| zhipeng | so many of this kind of scheduling will happen under the hood for cyborg | 16:06 |
| zhipeng | unless the driver is able to report to cyborg | 16:06 |
| zhipeng | #endmeeting | 16:14 |
| openstack | Meeting ended Wed Jun 28 16:14:16 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:14 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-06-28-15.13.html | 16:14 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-06-28-15.13.txt | 16:14 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg/2017/openstack_cyborg.2017-06-28-15.13.log.html | 16:14 |
| zhipeng | just in case ... | 16:14 |
| *** zhipeng has quit IRC | 16:32 | |
| *** joseppc has quit IRC | 16:37 | |
| *** jkilpatr has quit IRC | 16:50 | |
| *** jkilpatr has joined #openstack-cyborg | 18:09 | |
| *** NokMikeR has quit IRC | 18:14 | |
| *** mikeH has quit IRC | 19:36 | |
| *** mikeH has joined #openstack-cyborg | 21:10 | |
| *** mikeH has quit IRC | 21:49 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!