Comments (3)
Thanks. We are glad that you are interested in this project. I'm also doing some work on managing and scheduling accelerator (like GPU/MIG/vGPU/NPU) in Kubernetes. These are very import in Batch System. I'm looking forward to working together in the future to move Kueue forward.
Going further if we're doing multi-node with MPI and such, we need to think also about network topologies and node interconnects.
Kueue is currently not node-aware. I'm not sure it's possible to do some things like topologies and node interconnects. This feels more like the work to do in the scheduler. Can you share more details of your idea? Thanks
from kueue.
Correct, this feature request fits more in kube-scheduler and kubelet.
The perfect venue to discuss these ideas is the wg-batch https://github.com/kubernetes/community/tree/master/wg-batch
Hopefully we will set up a meeting today.
In the meantime, better open an issue in github.com/kubernetes/kubernetes
/close
from kueue.
@alculquicondor: Closing this issue.
In response to this:
Correct, this discussion fits more in kube-scheduler and kubelet.
The perfect venue to discuss these ideas is the wg-batch https://github.com/kubernetes/community/tree/master/wg-batch
Hopefully we will set up a meeting today.
In the meantime, better open an issue in github.com/kubernetes/kubernetes
/close
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.
from kueue.
Related Issues (20)
- Use k8s v1.30.0 HOT 4
- [WaitForPodsReady] The default configuration for `requeuingStrategy` is impractical HOT 9
- [WaitForPodsReady] There is no event for eviction on the workload
- [Flaky] Preemption In a cohort with StrictFIFO Should reclaim from cohort even if another CQ has pending workloads HOT 1
- [MultiKueue] Report ClusterQueue as inactive (misconfigured) if not applied to all flavors HOT 3
- [MultiKueue] Default managedBy for ClusterQueues configured to use MultiKueue AC HOT 12
- [MultiKueue] Report a ClusterQueue as inactive (misconfigured) if there is ProvReq used with MK HOT 2
- Kubernetes 1.30: kueue controller fails to remove scheduling gates HOT 16
- Changes to API comments aren't regenerated by generate-apiref HOT 3
- Fix transitions of Requeued condition HOT 3
- Fix logging of the workload status when using admission checks HOT 1
- Support all ProvisioningRequest's conditions
- [MultiKueue] e2e test fails occassionally HOT 2
- Add the ProvisioningRequest's classname annotation to pods HOT 4
- Ungating pods should use patch/apply instead of update HOT 4
- Cleanup creation of conditions HOT 1
- Scalability test is flaky HOT 5
- Support for sidecar containers HOT 2
- makefile cant exists gsed on MacOS HOT 1
- Single cluster e2e tests for pods fail to start occasionally
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from kueue.