Wednesday, 2019-05-15

*** Li_Liu has joined #openstack-cyborg01:48
*** wangzhh has joined #openstack-cyborg02:37
*** Sundar has joined #openstack-cyborg02:53
SundarCyborg weekly IRC call in 8 min02:54
SundarMake it 6 min02:54
*** xinranwang has joined #openstack-cyborg02:58
SundarHi Xinran02:59
xinranwangHi Sundar03:00
SundarWill wait for 2 min for folks to join03:00
Li_LiuHI guys03:00
SundarHi Li03:00
Li_LiuI am in China right now03:00
wangzhhHi all.03:01
SundarHi wangzhh03:01
SundarCoco said she can't make it. Let's get started then03:01
Sundar#startmeeting openstack-cyborg03:01
openstackMeeting started Wed May 15 03:01:43 2019 UTC and is due to finish in 60 minutes.  The chair is Sundar. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.03:01
*** openstack changes topic to " (Meeting topic: openstack-cyborg)"03:01
openstackThe meeting name has been set to 'openstack_cyborg'03:01
Sundar#topic Roll call03:01
*** openstack changes topic to "Roll call (Meeting topic: openstack-cyborg)"03:01
Li_Liu#info Li_Liu03:01
*** ikuo_o has joined #openstack-cyborg03:01
wangzhh#info wangzhh03:02
xinranwang#info xinranwang03:02
Sundar#info Sundar03:02
ikuo_o#info ikuo_o03:02
wangzhhHi ikuo_o.03:02
ikuo_oHi wangzhh.03:02
SundarPlease see agenda for today at:
SundarDoes anybody have anything else to add to it?03:02
ikuo_oThanks Sundar.03:03
*** Yumeng has joined #openstack-cyborg03:03
Sundar#topic Cyborg specs reviews03:03
*** openstack changes topic to "Cyborg specs reviews (Meeting topic: openstack-cyborg)"03:03
Yumeng#info Yumeng03:04
YumengHi all03:04
SundarHi Yumeng03:04
wangzhhMorning Yumeng.03:04
SundarTop 4 specs in:
SundarCan we get these 4 reviewed over the next few days?03:05
ikuo_oI and my colleagues are checking "Specification for Cyborg APIs Version 2".03:05
Li_LiuI will try my best03:05
SundarThanks, ikuo_o, Li_Liu.03:05
SundarGreat. Thanks wangzhh03:06
Sundar#topic Nova spec review03:06
*** openstack changes topic to "Nova spec review (Meeting topic: openstack-cyborg)"03:06
SundarLots of discussion in
Li_Liua lot of comments03:07
SundarDoes Huawei want a standard accelerator name for Ascend?03:07
SundarOnly FPGA and GPU for now03:07
Li_LiuAscend is on our agenda03:07
Li_LiuDoes it need to be mentioned in the nova spec?03:08
SundarLi_iu: Good. Just to be sure, instead of CUSTOM_ASCEND, you want ASCNED to be added to os-resource-classes, right?03:08
Li_Liuyou can check this with Yikun03:08
SundarYes, efried has said that explicitly.03:08
yikun#info Yikun03:09
SundarHi Yikun, just in time :)03:09
yikunsorry for late03:09
SundarNP. For Ascend, can we use a custom resource class name like CUSTOM_ASCEND? Or do you want to standardize it? We are standardizing only FPGA and GPU as of now03:10
wangzhhIs 'Ascend' popular  like GPU or FPGA?03:11
ikuo_oI don't know much about ascend...03:12
yikunFPGA GPU is the device type, but ascend maybe is not a standard type just a series, so I think maybe ascend should be cover in some thing like TPU_ASCEND or something03:12
SundarPerhaps it is best to leave it as a custom RC for now.03:12
Li_LiuTPU is very google specific03:12
yikunsorry, NPU03:13
SundarIf we propose a standard name, it should be added to the spec, justified and also proposed as a Nova patch.03:13
Sundaryikun, I believe you have an AI chip and a security-related accelerator?03:13
SundarNPU can also mean Network processing Unit. NNP = neural network processor03:14
yikunfor now, only AI chip, security-related accelerator comes from intel qat.03:14
Sundaryikun: OK. You can think about it. I am hoping to converge the Nova spec by May end, while we still have the momentum and attention.03:15
yikunSundar: ok, sure03:16
Sundar#topic Pilot merge plan03:16
*** openstack changes topic to "Pilot merge plan (Meeting topic: openstack-cyborg)"03:16
SundarI understand some of you find the merged patch to be very large to review. Is that the reason why we are not getting many reviews?03:16
SundarAnother way of asking: can you all review the code soon, or suggest ways to get the reviews done?03:18
Li_Liubtw, regarding changing from polling to event driven between nova and Cyborg, I just modify based on the pilot branch right?03:19
Li_Liuor submit a seperated patch?03:19
SundarLi_Liu: I am using the pilot code for various demos from now till mid-June. It will be best not to disrupt the pilot patches.03:20
wangzhhI'm fine with big patch or smaller one.03:20
Li_LiuSundar, sure, I will keep it clean in a different patch03:21
SundarLi_Liu and all: I am bit worried about what Nova guys are suggesting: there is no quantitative data on how it will affect Cyborg, right? FPGA programming can take a few seconds, and we are reducing that window of wait if we move the bind to the compute node03:21
Li_Liufor now I will try to review the spec and pilot branch03:21
wangzhhLi_liu, submit a seperated patch is OK. And pls add sundar's commit as a dependence.03:21
SundarWe should try to get some idea of how much the window gets reduced in different scenarios, like launch 1 VM, launch 2 VMs together, etc.03:22
SundarThe original suggestion at Stein PTG to do it this way was for multiple VM case. It was from Dan Smith, but he hasn't weighed in on thhe current round of dicussion03:23
Li_LiuAre they blocking the spec merging?03:24
Yumeng:Sundar, we only have 3 patches now, right?
Yumengif no more objections , we can merge asap03:25
SundarWell, John is saying that we either do it the way he suggests or justify doing it the current way. I don't have data either way.03:25
SundarLi_Liu: will you be putting a Nova patch too?03:26
Li_Liuthere will be changes in NOVA for sure03:26
SundarYou can submit that as a separate patch too. I can try to do some measurements in a live system with the combined set before we decide fully.03:27
SundarYumeng: It is really just .03:28
wangzhhYikun and I still have some comments.03:29
Sundarwangzhh: Good. Please go ahead with the comments.03:30
SundarI will also add some UT.03:30
SundarAlso, the pilot code makes some changes to the db schema. Just raising that point, if it wasn't clear.03:31
SundarNetworked FPGAs may make further changes but, unfortunately, I don't have full clarity on that yet.03:32
SundarBTW, I found this doc on db migration:
SundarIt talks of migration with expansion, contraction, etc. We aren't doing that, right?03:33
xinranwangBtw, I trying to figure out placement report things,  thers is one concern which is, the conductor will only receive the driver_device object with all infos insides. So conductor need to parse it to get the deployable, attribute etc. is it reasonable?03:34
Sundarxinranwang: The driver ovo is well-defined, so conductor should be able to get to the inner fields.03:35
wangzhhSundar, migration script supports some of these feature.03:35
SundarOk, good03:36
xinranwangyes, it should be. No info missing, but I just found it's a little complex for conductor03:36
xinranwanganyway, i have done like that. I will submit a patch soon, please review it if you got time03:36
Li_Liusounds good03:37
Sundar#action All to review the Cyborg specs, Nova spec and Cyborg pilot code (as much as possible) over the next week or so03:38
Sundar#topic Python 3 migration03:38
*** openstack changes topic to "Python 3 migration (Meeting topic: openstack-cyborg)"03:38
SundarBringing this up because I don;t know how much change this will need. SO may be it is best to discuss early.03:39
yikunSundar: [for the question in migrations script] the migration with expansion, contraction doc is just talk some rule, you can take any other project db migrations as example, we already have some script to complete add change filed in db.03:39
Sundaryikun: Good.03:40
Sundarikuo_o: Glad that NTT has taken up Python 3 migration. Do you need any help in doing some investigation to see how much change is needed? (You don;t need to make the changes just yet.)03:40
Li_LiuYikun, so it will be just the matter of running a script?03:40
ikuo_oSundar: Yes, now my colleagues are checking the code, and there are some questions.03:41
yikunLi_Liu: I think so03:41
ikuo_oI will send them later.03:41
Sundarikuo_o: Great, thanks03:41
Sundar#topic AoB03:42
*** openstack changes topic to "AoB (Meeting topic: openstack-cyborg)"03:42
SundarAnything else for today?03:42
Li_LiuDo we have zoom meeting tmr?03:43
SundarLi_Liu: No. I am actually attending a conference this week (even today). Besides, I don;t have any specific agenda for tomorrow. Do you or others have any?03:43
Li_Liuwhich conf are you going?03:44
SundarOpen Source Technical Symposium03:44
Li_Liucool :)03:45
SundarThis is conference season :) More coming up in June03:45
SundarThanks for a productive meeting, folks :)03:45
ikuo_oThanks sundar. I will invite my colleague for next IRC meeting!03:45
SundarSure, ikuo_o.03:46
SundarHave a good day, everybody!  (It is day for Li_Liu too :) )03:46
Li_LiuThank you Sundar :P03:46
ikuo_oHave a good day, all!03:46
wangzhhBye, all .03:46
*** openstack changes topic to "Pending patches (Meeting topic: openstack-cyborg)"03:46
openstackMeeting ended Wed May 15 03:46:47 2019 UTC.  Information about MeetBot at . (v 0.1.4)03:46
openstackMinutes (text):
*** Sundar has quit IRC03:47
*** Li_Liu has quit IRC03:51
*** ikuo_o has quit IRC03:55
*** links has joined #openstack-cyborg03:58
*** wangzhh has quit IRC05:57
*** xinranwang has quit IRC06:08
*** links has quit IRC06:38
*** links has joined #openstack-cyborg06:39
*** helenafm has joined #openstack-cyborg07:18
*** openstackgerrit has joined #openstack-cyborg08:13
openstackgerritXinran WANG proposed openstack/cyborg master: Improve Intel sysinfo based FPGA driver
openstackgerritXinran WANG proposed openstack/cyborg master: WIP: Placement report
openstackgerritXinran WANG proposed openstack/cyborg master: Improve Intel sysinfo based FPGA driver
openstackgerritXinran WANG proposed openstack/cyborg master: Improve Intel sysinfo based FPGA driver
*** links has quit IRC12:57
openstackgerritShaoHe Feng proposed openstack/cyborg master: bug fix: deploy every cyborg components correctly
*** links has joined #openstack-cyborg14:20
*** links has quit IRC15:26
*** helenafm has quit IRC15:33
*** links has joined #openstack-cyborg18:34
*** links has quit IRC19:07

Generated by 2.15.3 by Marius Gedminas - find it at!