Chunxiang Xu avadesian
Public Activity Repositories Starred Repositories
22
 Following
1
 Followers
11
 Achievement Badge
1
Loading Heatmap…

avadesian synced commits to simple-ray at avadesian/skypilot from mirror

4 hours ago

avadesian synced commits to resource-mgmt-pool-with-job at avadesian/skypilot from mirror

  • e276148daf Merge remote-tracking branch 'origin/master' into resource-mgmt-pool-with-job
  • 713daac964 [Catalog] Revert catalog change for Trainium and Inferentia chips (#7939) Revert catalog change
  • 179cfcea00 [Test]Production large load test (#7912) * readme update * test script * support clean up * production load test * longer sky status seconds * wait for termination * robust tests * longer timeout and sql update * cancel all jobs * sky down * clean up
  • 85623d818e [Tests] Increase Nightly Build Timeouts (#7933) Increase timeout to reduce false negatives.
  • 276999004e Updated for better Trainium/Inferentia support (#7896) * updated fetch_aws script to properly add Trainium/Inferentia to acclerator name and populate GpuInfo field. Added refresh of neuron based AMI IDs based on SSM parameters * removed incorrect code comment for generating vms.csv file * removed duplicate import * removed aws from .gitignore; fixed linting errors of fetch_aws.py file; fixed check of acc_name to be prefix rather than exact match of Trainium/Inferentia * fixed pylint formatting for aws.py * Update .gitignore --------- Co-authored-by: DanielZhangQD <36026334+DanielZhangQD@users.noreply.github.com>
  • Compare 259 commits »

4 hours ago

avadesian synced commits to pools-example-simple at avadesian/skypilot from mirror

4 hours ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • d1f0cdd25a Seeweb - Docker Images + Pinned Library (#7493) * seeweb-commit-review-11 * seeweb-commit-review-12 * seeweb-commit-review-13 * seeweb-commit-review-14 * Revert pytest.yml * Revert smoke-tests-trigger.yaml * Revert and fix test * Update seeweb accelerator test * Update test_sky_serve * Update test_sky_serve.py * seeweb-commit-review-15 * seeweb-commit-docs-01 * seeweb-commit-docs-02 * seeweb-commit-snapshotfix-01 * seeweb-commit-downfix-01 * seeweb-commit-docs-02 * seeweb-commit-libraryupd-01 * seeweb-commit-update_library-02 * seeweb-dockerimg-01 * fixeddependencies.py --------- Co-authored-by: Marco Cristofanilli <m4oc@users.noreply.github.com>
  • 0d92d4c909 [Pools] Fail Fast when Pools Misconfigured (#7930) * Error when pool misconfigured. * Update and format. * Don't error during num workers update. * Add check in sdk and unit test.
  • 713daac964 [Catalog] Revert catalog change for Trainium and Inferentia chips (#7939) Revert catalog change
  • Compare 3 commits »

4 hours ago

avadesian synced commits to lloyd/improve-concurrent-job-launch at avadesian/skypilot from mirror

4 hours ago

avadesian synced commits to lloyd/fix-pools-docs at avadesian/skypilot from mirror

4 hours ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • 179cfcea00 [Test]Production large load test (#7912) * readme update * test script * support clean up * production load test * longer sky status seconds * wait for termination * robust tests * longer timeout and sql update * cancel all jobs * sky down * clean up
  • 85623d818e [Tests] Increase Nightly Build Timeouts (#7933) Increase timeout to reduce false negatives.
  • 276999004e Updated for better Trainium/Inferentia support (#7896) * updated fetch_aws script to properly add Trainium/Inferentia to acclerator name and populate GpuInfo field. Added refresh of neuron based AMI IDs based on SSM parameters * removed incorrect code comment for generating vms.csv file * removed duplicate import * removed aws from .gitignore; fixed linting errors of fetch_aws.py file; fixed check of acc_name to be prefix rather than exact match of Trainium/Inferentia * fixed pylint formatting for aws.py * Update .gitignore --------- Co-authored-by: DanielZhangQD <36026334+DanielZhangQD@users.noreply.github.com>
  • af58b86002 Example policy for static GPU quota (#7917) gpu quota policy
  • c376dcff4b [Dashboard] enable filtering clusters and jobs by user (#7929) enable filtering clusters and jobs by user
  • Compare 14 commits »

17 hours ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • da450a0035 [Azure] Pin azcopy version (#7909) * [Azure] Pin azcopy version * check arm64 * fix lint --------- Co-authored-by: ZePing Guo <zp0int@qq.com>

2 days ago

avadesian synced commits to improve-tunnel-open at avadesian/skypilot from mirror

2 days ago

avadesian synced commits to admin-docs-update-2 at avadesian/skypilot from mirror

2 days ago

avadesian synced commits to 2904 at avadesian/skypilot from mirror

2 days ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • 11c5da60e6 [Catalog] Case insensitive check for canonicalized acc and cloud catalog (#7910) * [Catalog] Case insensitive check for canonicalized acc and instance type * minor
  • 44bb11e9e8 [Deployment] Use recreate for grafana deployment to avoid PVC blocking upgrade (#7904) Use recreate for grafana deployment to avoid RW-once PVC causing upgrade failure
  • 98b9165735 [Dev] Fix RTD call additional newlines
  • 277d04dcb5 [Dev] Trigger RTD version sync before build
  • dec33a9702 [Dev] Fix RTD project name
  • Compare 8 commits »

2 days ago

avadesian synced commits to batch-inference-with-pool at avadesian/skypilot from mirror

  • 641d2dcce3 Merge branch 'master' of github.com:skypilot-org/skypilot into batch-inference-with-pool
  • 7cff73216d [RunPod] Update runpod default image (#7906)
  • 8d17955d0b [SSH Node Pools] Allow per context config and provision_timeout (#7660) * Allow per context config for SSH node pool * fix the context config retrieval * strip ssh- instead * Add tests * fix tests * fix test * use a subset * Add ports back * better comment to explain ssh schema * Remove kueue and support provision_timeout * add comment * Add unit tests * same order * format * fix unit tests * format * Update sky/utils/schemas.py * fix constant * Add more tests --------- Co-authored-by: Daniel Shin <kyuseung1016@gmail.com> Co-authored-by: Daniel Shin <88547237+kyuds@users.noreply.github.com>
  • d8034a53bb [Core] Add support for num_gpus in setup. (#7092) * Add support for num_gpus in setup. * Add smoke test for regular launch. * Fix cluster name. * Format. * Change variable. * Change pools test. * Add note in docs. * Fix fractional GPUs. * Revert to old logic for runtime. * Add multiple accelerator example. * Add new unit test for multiple resources. * Rebase. * Format. * Get num_gpus from launched resources. * Revert "Get num_gpus from launched resources." This reverts commit cb5785ffd7a7a74a1a858f11a63c63123141a097. * Fix docs. * Remove unnecessary calls. * Remove resources fit. * Fix unit test. * format * format. * Remove extra test. --------- Co-authored-by: lloydbrownjr <lloydbrown@berkeley.edu>
  • 1337c50037 example rate limiter policy (#7902) * example rate limiter policy * docs * reasonable defaults * remove debug comment
  • Compare 627 commits »

2 days ago

avadesian synced commits to add-manual-rtd-build-action at avadesian/skypilot from mirror

2 days ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • 7cff73216d [RunPod] Update runpod default image (#7906)
  • 8d17955d0b [SSH Node Pools] Allow per context config and provision_timeout (#7660) * Allow per context config for SSH node pool * fix the context config retrieval * strip ssh- instead * Add tests * fix tests * fix test * use a subset * Add ports back * better comment to explain ssh schema * Remove kueue and support provision_timeout * add comment * Add unit tests * same order * format * fix unit tests * format * Update sky/utils/schemas.py * fix constant * Add more tests --------- Co-authored-by: Daniel Shin <kyuseung1016@gmail.com> Co-authored-by: Daniel Shin <88547237+kyuds@users.noreply.github.com>
  • Compare 2 commits »

3 days ago

avadesian synced commits to allow-per-context-config-for-ssh at avadesian/skypilot from mirror

  • 9093c8c658 Merge branch 'master' into allow-per-context-config-for-ssh
  • d8034a53bb [Core] Add support for num_gpus in setup. (#7092) * Add support for num_gpus in setup. * Add smoke test for regular launch. * Fix cluster name. * Format. * Change variable. * Change pools test. * Add note in docs. * Fix fractional GPUs. * Revert to old logic for runtime. * Add multiple accelerator example. * Add new unit test for multiple resources. * Rebase. * Format. * Get num_gpus from launched resources. * Revert "Get num_gpus from launched resources." This reverts commit cb5785ffd7a7a74a1a858f11a63c63123141a097. * Fix docs. * Remove unnecessary calls. * Remove resources fit. * Fix unit test. * format * format. * Remove extra test. --------- Co-authored-by: lloydbrownjr <lloydbrown@berkeley.edu>
  • 1337c50037 example rate limiter policy (#7902) * example rate limiter policy * docs * reasonable defaults * remove debug comment
  • 3d58efeee2 [Core] make run ray status during refresh safe from referencing unassigned local variable (#7901) make run ray status to check ray cluster healthy safe from local variable referenced before assignment
  • 3e5c599fde [Logs] make provision logs not timeout (#7888) * make provision logs not timeout * fix indenting issue * simplify worker logic
  • Compare 145 commits »

3 days ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • d8034a53bb [Core] Add support for num_gpus in setup. (#7092) * Add support for num_gpus in setup. * Add smoke test for regular launch. * Fix cluster name. * Format. * Change variable. * Change pools test. * Add note in docs. * Fix fractional GPUs. * Revert to old logic for runtime. * Add multiple accelerator example. * Add new unit test for multiple resources. * Rebase. * Format. * Get num_gpus from launched resources. * Revert "Get num_gpus from launched resources." This reverts commit cb5785ffd7a7a74a1a858f11a63c63123141a097. * Fix docs. * Remove unnecessary calls. * Remove resources fit. * Fix unit test. * format * format. * Remove extra test. --------- Co-authored-by: lloydbrownjr <lloydbrown@berkeley.edu>
  • 1337c50037 example rate limiter policy (#7902) * example rate limiter policy * docs * reasonable defaults * remove debug comment
  • 3d58efeee2 [Core] make run ray status during refresh safe from referencing unassigned local variable (#7901) make run ray status to check ray cluster healthy safe from local variable referenced before assignment
  • 3e5c599fde [Logs] make provision logs not timeout (#7888) * make provision logs not timeout * fix indenting issue * simplify worker logic
  • 4fb702215a [Dashboard] Show Succeeded Pools in Dropdown (#7855) Show succeeded pools.
  • Compare 11 commits »

4 days ago

avadesian synced commits to lloyd/pools-run-message at avadesian/skypilot from mirror

4 days ago

avadesian synced commits to serve-files-stash-to-db at avadesian/skypilot from mirror

5 days ago

avadesian synced commits to master at avadesian/skypilot from mirror

  • 767213dc7c [k8s] retry k8s api list namespaced pod on empty response (#7854) * add more debug logs around k8s list_namespaced_pod * label selector * [k8s] retry k8s api list namespaced pod on empty response * tests * reorder retry logic * only retry for status refresh daemon and status -r * remove unnecessary retry_if_missing * remove defaults * fix test * update retry if missing logic to only run on sky status -r * clarity * change retry if missing logic for refresh * update logic * update docstring * only kubernetes docstring * adjust logic and add comment * tighten * update test
  • d45fd6330a [Core] Ensure SSH tunnel process is terminated when cluster is torn down (#7887) * Add test_no_ssh_tunnel_process_leak_after_teardown * close skylet ssh tunnel in teardown_no_lock * use kill_children_processes instead * only update db if stop * Revert "only update db if stop" This reverts commit fe96ee18be8cee583f68356b3436a9b701e64ea2. * graceful termination * proc.pid
  • 38c27f5ea6 [Kubernetes] Check supported features for all allowed contexts (#7878) * check supported features for all contexts * add ut * check features support n parallel
  • e07d94d3b5 [Dashboard] Return resources_str_full for cluster records of the dashboard requests (#7880) return resources_str_full for cluster records of the dashboard requests
  • 75458afd67 Admin policy docs update (#7866) * init * init2 * minor correction --------- Co-authored-by: tv <tv@tvs-MacBook-Pro.attlocal.net>
  • Compare 9 commits »

5 days ago

Please read the following content carefully:

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

Agree and continue
Disagree, exit