| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-cyborg - 2018-05-28 | |||
| 14:29:42 | Li_Liu | I will update another patch to address Rushil's comments | |
| 14:29:48 | Li_Liu | in term of code dev | |
| 14:29:55 | Li_Liu | I have started some of it | |
| 14:30:53 | shaohe_feng | #link https://review.openstack.org/#/c/559395/ | |
| 14:31:08 | shaohe_feng | yes, Rushil give some comments. | |
| 14:31:25 | shaohe_feng | you need to address them. | |
| 14:31:32 | Li_Liu | yup | |
| 14:31:50 | shaohe_feng | Li_Liu, I will help sundar on his nova-cyborg integration focuses on placement provider update. | |
| 14:32:11 | Li_Liu | ok | |
| 14:32:32 | shaohe_feng | Li_Liu, which comment in cyborg will update the placement, conductor or agent? | |
| 14:33:57 | shaohe_feng | I remeber nova update the placement in n-cpu (same with cyborg agent) | |
| 14:34:01 | Li_Liu | I think agent for sure needs to update placement | |
| 14:34:19 | shaohe_feng | I think so. | |
| 14:35:21 | shaohe_feng | This is NOVA BP for nest provider | |
| 14:35:22 | shaohe_feng | #link https://blueprints.launchpad.net/nova/+spec/nested-resource-providers | |
| 14:35:30 | Li_Liu | in terms of conductor, I am not sure, but I think Cyborg controller needs to update placement | |
| 14:37:45 | shaohe_feng | Li_Liu, I found you have add placement client in Cyborg | |
| 14:38:16 | Li_Liu | yes | |
| 14:38:16 | shaohe_feng | Guess it does not support nest provider, right? | |
| 14:38:35 | Li_Liu | at that time, prob not | |
| 14:38:40 | shaohe_feng | Nova's nest provider patch has been merged. | |
| 14:39:01 | Li_Liu | sure, I will see what I can do about it | |
| 14:39:29 | shaohe_feng | So in cyborg, we can support nest provider. | |
| 14:39:45 | Li_Liu | yup, we should | |
| 14:40:13 | shaohe_feng | and last IRC meeting, we have agreement to support multi-resource classes. | |
| 14:40:30 | shaohe_feng | nest provider + multi-resource classes | |
| 14:40:43 | Li_Liu | ok | |
| 14:41:31 | shaohe_feng | Ok, I think that's all from me. | |
| 14:42:06 | shaohe_feng | Li_Liu, anything more need to discuss or update? | |
| 14:42:16 | Li_Liu | a lot things need to work on | |
| 14:42:22 | shaohe_feng | yes. | |
| 14:42:41 | shaohe_feng | What's else can I help, let me know. | |
| 14:42:58 | Li_Liu | sure, will let you know in case I need help | |
| 14:44:01 | shaohe_feng | OK, let's work together to push cybork keep going | |
| 14:44:51 | Li_Liu | I will try to give a summary on the summit during the project general irc this week | |
| 14:45:22 | shaohe_feng | good, expect it. Thanks. | |
| 14:46:05 | shaohe_feng | If no more, we can end the meeting. | |
| 14:49:36 | shaohe_feng | #endmeeting | |
| 14:49:38 | openstack | Meeting ended Mon May 28 14:49:36 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | |
| 14:49:39 | openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_cyborg_driver/2018/openstack_cyborg_driver.2018-05-28-14.02.html | |
| 14:49:40 | openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_cyborg_driver/2018/openstack_cyborg_driver.2018-05-28-14.02.txt | |
| 14:49:41 | openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_cyborg_driver/2018/openstack_cyborg_driver.2018-05-28-14.02.log.html | |
| 15:04:14 | openstackgerrit | Li Liu proposed openstack/cyborg master: Added cyborg fpga programming spec https://review.openstack.org/559395 | |
| #openstack-cyborg - 2018-05-29 | |||
| 12:50:21 | openstackgerrit | wangzhh proposed openstack/cyborg master: Load cyborg-api app with paste_deploy https://review.openstack.org/570931 | |
| 16:34:46 | openstackgerrit | ShaoHe Feng proposed openstack/cyborg master: doc fix: devstack setup doc can not display well https://review.openstack.org/570968 | |
| 17:07:00 | openstackgerrit | Li Liu proposed openstack/cyborg master: Added cyborg fpga programming spec https://review.openstack.org/559395 | |
| #openstack-cyborg - 2018-05-30 | |||
| 02:11:27 | openstackgerrit | Li Liu proposed openstack/cyborg master: Added cyborg fpga programming spec https://review.openstack.org/559395 | |
| 07:59:04 | shaohe_feng | zhipengh[m], I have update the wiki https://wiki.openstack.org/wiki/Cyborg#How_to_contribute | |
| 07:59:33 | shaohe_feng | add DevStack Quick Start link to the wiki | |
| 08:00:25 | shaohe_feng | so New contributors can follow it setup their cyborg env. | |
| 08:01:14 | shaohe_feng | Any issue during the setup, they can fix it and feedback to community. | |
| 09:24:11 | openstackgerrit | Merged openstack/cyborg master: doc fix: devstack setup doc can not display well https://review.openstack.org/570968 | |
| 09:25:50 | openstackgerrit | wangzhh proposed openstack/cyborg master: Load cyborg-api app with paste_deploy https://review.openstack.org/570931 | |
| 09:39:25 | openstackgerrit | wangzhh proposed openstack/cyborg master: Load cyborg-api app with paste_deploy https://review.openstack.org/570931 | |
| 14:03:08 | Sundar | #info Sundar | |
| 14:09:30 | sum12 | no meeting today ? | |
| 14:10:10 | Li_Liu | still waiting for Howard | |
| 14:10:44 | Li_Liu | shaohe_feng can you start the meeting for us? | |
| 14:10:56 | Li_Liu | in case Howard is in middle of something\ | |
| 14:11:28 | Sundar | The meeting topic is set to 'spec review day'. Was that announced earlier? | |
| 14:12:19 | Li_Liu | I think that is for a few week earlier | |
| 14:14:19 | shaohe_feng | Li_Liu, hello | |
| 14:15:12 | Li_Liu | let start the meeting shaohe_feng | |
| 14:15:46 | shaohe_feng | #startmeeting openstack-cyborg | |
| 14:16:02 | shaohe_feng | #topic Roll Call | |
| 14:16:16 | Sundar | #info Sundar | |
| 14:16:17 | shaohe_feng | #info shaohe_feng | |
| 14:16:39 | Li_Liu | #info Li_Liu | |
| 14:17:31 | Li_Liu | #topic summit summary | |
| 14:17:39 | Li_Liu | am I doing it right? | |
| 14:17:55 | Li_Liu | changing the topic | |
| 14:18:02 | shaohe_feng | yes. | |
| 14:18:33 | shaohe_feng | welcome Li_Liu introduce his experience during summit. | |
| 14:18:58 | Li_Liu | ok, a few of us went to the summit last week. Welcome to Canada for those of you who came | |
| 14:19:30 | Li_Liu | We had a few talks given regarding to Cyborg | |
| 14:20:22 | Li_Liu | I gave a Cyborg project update. Friends from Lenovo gave a demo using Cyborg to manage GPUs in their ThinkCloud product | |
| 14:20:56 | Li_Liu | JF. Ding also gave a talk with 99 cloud folks, which involves cyborg a lot | |
| 14:21:29 | Li_Liu | the general feedback is great. A lot of attentions are coming the way to Cyborg | |
| 14:21:40 | shaohe_feng | good. | |
| 14:22:06 | Li_Liu | couple quick comments from the audiences | |
| 14:22:40 | sum12 | any recurring comments ? | |
| 14:23:23 | Li_Liu | 1. Managing GPUs is still not a killer app for Cyborg. Prob showing other capabilities on top of this might draw more attentions | |
| 14:25:12 | Li_Liu | 2. In order to become a strong piece of OpenStack or even addition to k8s in the future, we need to push the development further and faster | |
| 14:25:36 | Li_Liu | as we are actually facing competitions from the Nova vGPU management. | |
| 14:26:09 | Li_Liu | currently we are at pretty much the same stage as them. | |
| 14:27:28 | Li_Liu | some people are looking at us since they think currently vGPU management in Nova should just be temperate. | |
| 14:28:14 | Li_Liu | They believe once Cyborg becomes more mature, it should be the one actually solve the general problem | |
| 14:28:32 | Li_Liu | This is from me. | |
| 14:28:41 | Li_Liu | anyone else has anything to share? | |
| 14:29:24 | Sundar | I moderated a session on Cyborg/FPGA in Cloud/NFV. | |
| 14:29:56 | shaohe_feng | Sundar, can you give a summary about it? | |
| 14:30:04 | Sundar | It was meant for discussion with operators and end users. We had Verizon, SKA and CERN there, apart from many developers, incl. Nova developers | |
| 14:30:35 | Sundar | The discusison is here: #link https://etherpad.openstack.org/p/Cyborg-FPGA-Support-for-Cloud-NFV | |
| 14:31:14 | Sundar | One of the main conclusions is that the model where VM issues programming requests (what we called run-time programming) is important | |
| 14:31:27 | Sundar | However, we do not have a good end-to-end flow for that today. | |
| 14:31:44 | Sundar | This is an open question that we should address. | |
| 14:32:11 | Li_Liu | The main concern is who to trigger the programming right? | |
| 14:32:57 | Sundar | That is one of the main concerns, the other being the validity of the bitstream itself | |
| 14:33:11 | Li_Liu | ah, that one | |
| 14:34:15 | Sundar | For example, if a developer is developing new RTL inside the VM, the resulting bitstream may have bugs. Can that compromise the shell logic? Can it result in local hot spots or other power issues that affect the FPGA chip itself? | |