Skip to content

GREP-375 add scheduler backend framework#372

Merged
unmarshall merged 47 commits into
ai-dynamo:mainfrom
kangclzjc:grep_scheduler_backend
Feb 20, 2026
Merged

GREP-375 add scheduler backend framework#372
unmarshall merged 47 commits into
ai-dynamo:mainfrom
kangclzjc:grep_scheduler_backend

Conversation

@kangclzjc

@kangclzjc kangclzjc commented Jan 27, 2026 •
edited by unmarshall
Loading

Copy link
Copy Markdown
Contributor

What type of PR is this?

/kind documentation

What this PR does / why we need it:

Add scheduler backend framework to support multiple scheduler backends

Which issue(s) this PR fixes:

Fixes #275
Fixes #375

Special notes for your reviewer:

Does this PR introduce a API change?

NONE

Additional documentation e.g., enhancement proposals, usage docs, etc.:

Introduce GREP for scheduler backend framework

@kangclzjc kangclzjc marked this pull request as ready for review January 27, 2026 12:45
@kangclzjc kangclzjc changed the title GREP add scheduler backend framework GREP-375 add scheduler backend framework Jan 28, 2026
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
sanjaychatterjee
sanjaychatterjee reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
@copy-pr-bot

copy-pr-bot Bot commented Feb 3, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 3, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Ronkahn21
Ronkahn21 reviewed Feb 3, 2026
View reviewed changes

@Ronkahn21 Ronkahn21 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Half way. I will complete the review tomorrow

Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md
Comment thread docs/proposals/375-scheduler-backend-framework/README.md
Ronkahn21
Ronkahn21 reviewed Feb 5, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
@kangclzjc kangclzjc force-pushed the grep_scheduler_backend branch 2 times, most recently from 7012497 to 3ac9845 Compare February 18, 2026 06:10
sanjaychatterjee
sanjaychatterjee reviewed Feb 18, 2026
View reviewed changes

@sanjaychatterjee sanjaychatterjee left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. Made a couple of minor suggestions to update the GREP. Thanks!

Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
unmarshall
unmarshall reviewed Feb 18, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
@kangclzjc kangclzjc force-pushed the grep_scheduler_backend branch from 16e1c09 to ef9a34e Compare February 19, 2026 01:24
unmarshall
unmarshall reviewed Feb 19, 2026
View reviewed changes
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md Outdated
Comment thread docs/proposals/375-scheduler-backend-framework/README.md
@kangclzjc kangclzjc force-pushed the grep_scheduler_backend branch 2 times, most recently from b6348bc to 1e4e042 Compare February 20, 2026 01:35
@kangclzjc @unmarshall
add scheduler backend grep
a485ae3 
Signed-off-by: kangclzjc <kangz@nvidia.com>
@kangclzjc @unmarshall
add pic
9ecb007 
Signed-off-by: kangclzjc <kangz@nvidia.com>
@kangclzjc @unmarshall
refine proposal
94114f5 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
update toc
595fa25 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
kangclzjc and others added 23 commits February 20, 2026 23:17
@kangclzjc @unmarshall
Apply suggestions from code review
f89ab7a 
remove phase1 in limitation

Co-authored-by: Madhav Bhargava <madhav.bhargava@sap.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
add schedulers capability mismatch
f550b35 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
format toc
b4e5188 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
refine podgang lifecycle
0bb47c8 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
remove dependencies
edd5f93 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
modify scheduler backend framework picture
c91b471 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
modify scheduler backend framework picture
67dfa17 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @Ronkahn21 @unmarshall
Update docs/proposals/375-scheduler-backend-framework/README.md
fb78f74 
Move scheduler string to struct

Co-authored-by: Ron Kahn <122778260+Ronkahn21@users.noreply.github.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
add new requirements
31fa925 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
refine config to schedulerConfig
401c426 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
Remove No Goas
44e0344
@kangclzjc @unmarshall
add validatepcs
e99e8fb 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
modify picture
b88c86f 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
replace user with admin
a73769e 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
refine schedulerName
abe353c 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
refine KAI name
d6b7caf 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
Refine description
cd98fd2 
Co-authored-by: Madhav Bhargava <madhav.bhargava@sap.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
Refine backend controller
42b4daa 
Co-authored-by: Madhav Bhargava <madhav.bhargava@sap.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
Update Podclique controller description
b44b8b5 
Co-authored-by: Madhav Bhargava <madhav.bhargava@sap.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
Update Kai backend description
2679b40 
Co-authored-by: Madhav Bhargava <madhav.bhargava@sap.com>
Signed-off-by: Kang Zhang <100667394+kangclzjc@users.noreply.github.com>
@kangclzjc @unmarshall
remove schedulerName in PCS level and refine phases
b20a02e 
Signed-off-by: Kang Zhang <kangz@nvidia.com>
@kangclzjc @unmarshall
clarify the gangscheduling config
532789f 
Signed-off-by: Kang Zhang <kangz@nvidia.com>

add non-goals

Signed-off-by: Kang Zhang <kangz@nvidia.com>

add e2e test config

Signed-off-by: Kang Zhang <kangz@nvidia.com>
@unmarshall
Refactored some sections in the GREP.
84a5388 
Signed-off-by: Madhav Bhargava <madhav.bhargava@sap.com>
@unmarshall unmarshall force-pushed the grep_scheduler_backend branch from 863cc0b to 84a5388 Compare February 20, 2026 17:48
@unmarshall unmarshall merged commit a92002c into ai-dynamo:main Feb 20, 2026
8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@unmarshall unmarshall unmarshall approved these changes

@sanjaychatterjee sanjaychatterjee sanjaychatterjee approved these changes

@gflarity gflarity Awaiting requested review from gflarity gflarity is a code owner

@shayasoolin shayasoolin Awaiting requested review from shayasoolin shayasoolin is a code owner

@Ronkahn21 Ronkahn21 Awaiting requested review from Ronkahn21 Ronkahn21 is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

GREP: add scheduler backend framework Add Native Support for Kubernetes Workload API to Enable Gang Scheduling

4 participants

@kangclzjc @unmarshall @sanjaychatterjee @Ronkahn21