Open Stack

Hi all.



I was going to attempt to enumerate some sort of grand vision for Cyborg and our priorities for the next cycle. As I thought about it, though, I realized

these remain largely the same as they always have been: provide a reliable, scalable, and efficient hardware-accelerated device management platform.

In the past, we have added support for GPU, FPGA, NVMe SSD, and will add support for AEP, SPDK and other devices in the future.



In order to increase the ability to interact with the VM, we were completed the feature of nova-cyborg interaction in Ussuri release. The exact details of

how we do that may vary somewhat from cycle to cycle, and I was completed the support of nova rebuild/evacuate server and shelve/unshelve in the last two

releases, which greatly improved the convenience of user operations. Such as the suspend and resume, the PoC code you can review https://review.opendev.org/c/openstack/nova/+/729945.

And we also tried to support the code migration in the future, of course that need a spec and to talk with nova team.



The log management project (Venus) team and I have completed the integration of Venus to Cyborg. We can use Venus to quickly retrieve Cyborg logs, shorten

the problem location time, etc., so as to improve the efficiency of the Cyborg project. In the future, I will have in-depth exchanges with the Venus

development team to continue to enhance this feature.



In these two releases, I have promoted many newcomers to join the Cyborg team (such as Wenping Song, Eric Xie, fossnqiu, Zhiguang Wang, Arthur Dayne etc.),

and they have made great contributions to the Cyborg project. In the future, I would like to help more people.



I hope to get your support, thanks.



Bailin Zhang (brinzhang)





brinzhang

Inspur Electronic Information Industry Co.,Ltd.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20210902/6af683fa/attachment.html>