Awesome.

Thanks Xinran.

 

¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª¡ª

Brin Zhang |  ÕÅ°ÙÁÖ

 

CBRD  | ÔƼÆËãÓë´óÊý¾ÝÑз¢²¿

 

T: 13203809727

 

E: zhangbailin@inspur.com

 

 

cid:image002.jpg@01D66424.D3B73D90

À˳±µç×ÓÐÅÏ¢²úÒµ¹É·ÝÓÐÏÞ¹«Ë¾

Inspur Electronic Information Industry Co.,Ltd.

Ö£ÖÝÊÐÖ£¶«ÐÂÇøÐÄâù·278ºÅ»ùÔËͶ×Ê´óÏÃ17²ã

17th Floor, , No.278 Xinyi Road, Zhengzhou City

À˳±Ôƺ£

 

 

·¢¼þÈË: Wang, Xin-ran [mailto:xin-ran.wang@intel.com]
·¢ËÍʱ¼ä: 2021Äê4ÔÂ25ÈÕ 17:56
ÊÕ¼þÈË: OpenStack Discuss <openstack-discuss@lists.openstack.org>
Ö÷Ìâ: [cyborg][ptg]Cyborg vPTG Summary April 2021

 

Hi all,

 

Please ignore my previous message, it is sent to wrong person. Sorry about this.  

 

Thanks for all your participation! We¡¯ve conducted a successful meeting last week.

 

Here is the aggregated summary from Cyborg vPTG discussion. Please check it out and feel free to feedback any concerns you might have.

 

We did a retrospective of Wallaby release, including:

    * We supports more operations supported for a VM with accelerator attached.

    * We introduced new drivers for Intel x710 NIC and Inspur¡¯s NVMe SSD Card.

    * We implemented a new configuration file allowing more flexible device configuration.

 

Topic discussion:

    Here's some major discussion and conclusion of Cyborg vPTG. For more details, please refer to the etherpad[1].

 

   * More nova operation supporting:

           - We prioritized the tasks: 1. suspend/resume. 2. cold migration. 3. live migration.

   * vGPU support:

           - We reached an internal agreement on whole workflow which can be apply as a generic framework for mdev device.

   * API enhancement:

           Some of the following items requires a new micro-version.

           -  Add refresh policy check for all APIs.

           -  Add device profile update API.

          - Add ARQ query by multiple instances API.

           -  Add disable/enable device API, this one requires a spec first.

           -  Enhance device profile show API with more information.

   * Cleanup issue:

           - This issue comes from the case where one compute node shutdown accidently, and the accelerator records

           in placement and cyborg DB remains as the orphaned resources. We agreed to implement a mechanism to clean

           up the orphaned resources, this one also need a spec.

 

[1] https://etherpad.opendev.org/p/cyborg-xena-ptg

 

 

Thanks,

Xin-Ran Wang