[Starlingx-discuss] StarlingX 3.0 features
Zhao, Forrest
forrest.zhao at intel.com
Wed Jul 3 01:57:32 UTC 2019
Yes, it’s a committed feature for STX 3.0 from networking team.
The latest patch set of its spec at https://review.opendev.org/#/c/655830/ has addressed all open comments. We’ll ping reviewers to move forward.
Thanks,
Forrest
From: Hu, Yong
Sent: Wednesday, July 3, 2019 9:15 AM
To: Zhao, Forrest <forrest.zhao at intel.com>; Jones, Bruce E <bruce.e.jones at intel.com>; starlingx-discuss at lists.starlingx.io
Subject: Re: [Starlingx-discuss] StarlingX 3.0 features
Hi Forrest,
Did you mean Network team has plan to commit this feature for Stx.3.0?
If so, are we going forward with this spec/patch: https://review.opendev.org/655830, in which the last comment was made on May 5th?
Regards,
Yong
On 02/07/2019, 1:42 AM, "Zhao, Forrest" <forrest.zhao at intel.com<mailto:forrest.zhao at intel.com>> wrote:
Hi Bruce and TSC reviewers,
Here are intents of “containerize OVS DPDK”,
1. As StarlingX moves to containerization, most OpenStack components have been containerized. That includes OVS containerization, but OVS-DPDK is still running on host. It’s better to containerize OVS/DPDK as well, to leverage benefits brought by containerization.
2. Currently, StarlingX supports OVS and OVS-DPDK. OVS is managed by openstack-helm, and running in container. But OVS-DPDK is managed by puppet and run directly on the host. Maintaining two implements and keeping them consistent cost more resources than maintaining just one implement. For example. If we want to make some changes(upgrade OVS version, enable some features), we need the changes at two places. It introduces much more upgrade/maintenance costs. “Containerize OVS DPDK” can eliminate such duplication and inconsistency.
Thanks,
Forrest
From: Jones, Bruce E [mailto:bruce.e.jones at intel.com]
Sent: Thursday, June 27, 2019 11:44 PM
To: starlingx-discuss at lists.starlingx.io<mailto:starlingx-discuss at lists.starlingx.io>
Subject: [Starlingx-discuss] StarlingX 3.0 features
Here is a list of StarlingX 3.0 features that our team plans to work on. I would like to ask the TSC to please review the “Not yet discussed” features before closing on the 3.0 feature list. Thank you!
brucej
Work item
TSC status
Lead
Status
IA platform features
Not yet discussed
Abraham, Saul, Ada
Most work is validation, new features are integrated as we adopt newer kernels over time. Real time features for industrial use cases are going into 5.x kernels and may (or may not) be back-portable.
Containerize OVS DPDK
AR Yong
Forrest
Not yet approved, Yong to get with Forrest and confirm intent
Performance testing
Not yet discussed
Victor, Ada
Proposal in progress
FPGA accelerator support
Push to 4.0
Abraham, Ada
Too big for 3.0 but will likely need to start soon. FPGA hardware has been ordered
OpenStack Train integration
Approved
Bruce (Dean)
Continuous integration from OpenStack master
Containerized Ceph
Push to 4.0
Vivian
Too big for 3.0 but will likely need to start soon
Time Sensitive Networking
Approved
Forrest
Spec in progress
Kubernetes plugins for IA
Partial
Cindy
Some reviews in progress, QAT approved, FPGA likely 4.0
Redfish
Approved
Cindy
Spec in progress
IOT device management
Not yet discussed
Abraham
Demo’d @ Denver. POC / pathfinding work for a customer in progress, item likely too big for 3.0
SUSE build support & enablement
Approved
Abraham, Saul
In progress, previously approved for 2.0 and continuing on
Containerized OpenStack Clients
Approved
Dean?
Nearly completed for 2.0, pushed to 3.0
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20190703/c775d9e8/attachment-0001.html>
More information about the Starlingx-discuss
mailing list