v0.2.0
skypilot-org/skypilotv0.2.0Oct 11, 2022by concretevitamin
Full Release Notes
We are excited to release SkyPilot 0.2.0, which receives a host of new features, with many enhancements and fixes. ## Highlights * [**Managed Spot**](https://skypilot.readthedocs.io/en/latest/examples/spot-jobs.html) is made much more robust and easier to use. * Try using `sky spot launch` on your existing yamls! * We've seen users running 1000s of spot jobs in a recurring schedule. * [**TPU Pods**](https://skypilot.readthedocs.io/en/latest/reference/tpu.html) are now supported. * To use a TPU Pod, simply modify e.g., `accelerators: tpu-v2-8` to `accelerators: tpu-v2-32`. * [**Benchmark**](https://skypilot.readthedocs.io/en/latest/reference/benchmark/index.html): use `sky bench` to easily measure the performance and cost of different cloud resources for your task. * Provisioning is sped up by ~1 minute. * Catalog is updated to [V3](https://github.com/skypilot-org/skypilot-catalog/tree/master/catalogs/v3) with 100s of resource changes and 1000s of price changes. * `A100-80GB` is now available on 3 clouds. Check out `sky show-gpus -a` for GPU prices. * No action needed as this will be automatically downloaded. ## CLI & Task interface New Features * Add zone support in YAML #1014 * Add shell completion support for CLI by #1162 * Add `--no-setup` option to `sky launch` to allow for remounting of files without running setup commands again #1184 * Add `sky start --all` to start all clusters #1065 * Add glob support for `sky storage delete` #1117 * Add `--no-follow` option to `sky logs` and `sky spot logs` (print logs so far and exit) Enhancements * Show vCPUs in optimizer/benchmark messages #1076 * Make entrypoint optional: for quick VM launching, no more `sky launch <flags> ''`, simply do `sky launch <flags>` #1191 * Make `sky check` automatically enable necessary GCP APIs (#1197, #1209); make it more robust for AWS checks (#1194) ## Managed spot New Features * `sky spot launch` now automatically translates file_mounts in a YAML to use cloud storage. #1081 #1215 * This means the same YAML for on-demand resources launched by `sky launch` can now be launched by `sky spot launch`. * Add `--retry-until-up` for `sky spot launch`; improve the responsiveness for `sky spot cancel` https://github.com/skypilot-org/skypilot/pull/1098 * Expose a `$SKYPILOT_RUN_ID` environment variable shared by all recoveries of the same spot job (useful for identifying it in Weights & Biases) #1196 * See the last Note block in [docs](https://skypilot.readthedocs.io/en/latest/examples/spot-jobs.html). Enhancements * Distinguish spot controller names for different users https://github.com/skypilot-org/skypilot/pull/1101 * **This may leak an old stopped controller** if you have used `spot launch` with <= 0.1.2. * Add retry for spot cluster termination https://github.com/skypilot-org/skypilot/pull/1139 * Enable purge for spot controller https://github.com/skypilot-org/skypilot/pull/1107 * Show FAILED_CONTROLLER when controller exit abnormally https://github.com/skypilot-org/skypilot/pull/1143 * Make get_job_timestamp fetching more robust https://github.com/skypilot-org/skypilot/pull/1148 * Fail early when spot cluster name too long occurs on GCP https://github.com/skypilot-org/skypilot/pull/1183 Fixes * Fix the retry logic for spot cluster launching https://github.com/skypilot-org/skypilot/pull/1150 * Fix non-persistent storage deletion for spot #974 * Fix spot recovery without cloud specified #1077 * Fix spot job duration https://github.com/skypilot-org/skypilot/pull/1104 * Fix `sky spot status -a` for resources and region information https://github.com/skypilot-org/skypilot/pull/1135 ## TPU support * Support TPU Pod https://github.com/skypilot-org/skypilot/pull/1001 ## Provisioner Enhancements * Improving provision speed by ~1 minute (#1092, #1103, #1108, #1111, #1126) * Add host VM - GPU compatibility checks for GCP https://github.com/skypilot-org/skypilot/pull/989 Fixes * Fix GCP VM leak issue https://github.com/skypilot-org/skypilot/pull/1102 * Fix GCP A100 launch error https://github.com/skypilot-org/skypilot/pull/1166 * Fix K80 gpunode by correcting GCP image version https://github.com/skypilot-org/skypilot/pull/1090 ## On-prem Enhancements * **Simplified on-prem deployment** * `sky admin deploy` now automatically installs `skypilot`, `ray` (and `python3` and `pip3`) on the local cluster under admin user #1116 * Add cluster config schema check https://github.com/skypilot-org/skypilot/pull/1044 * Modify Sky Admin's Setup on Docs https://github.com/skypilot-org/skypilot/pull/1085 * Align Python Versions https://github.com/skypilot-org/skypilot/pull/1086 Fixes * Fix Sky Status Logging https://github.com/skypilot-org/skypilot/pull/1041 ## Backend Enhancements * Catalog is updated to V3 with 100s of resource changes and 1000s of price changes #1204 * Canonicalize accelerator names in Resources https://github.com/skypilot-org/skypilot/pull/1075 * Reduce the frequency of job status update and remove parallel query https://github.com/skypilot-org/skypilot/pull/1096 * Increase thread limit and fix nofile limit https://github.com/skypilot-org/skypilot/pull/1128 Fixes * [Storage] Fix public bucket source check in SkyPilot Storage https://github.com/skypilot-org/skypilot/pull/1087 * Fixes ray dashboard hanging problem (#1088) https://github.com/skypilot-org/skypilot/pull/1109 * Fix placement group not scheduled issue (issue #1130) https://github.com/skypilot-org/skypilot/pull/1134 ## Misc. enhancements * **New example**: Stable Diffusion https://github.com/skypilot-org/skypilot/pull/1149 * `pip install skypilot` now installs `skypilot[aws]` by default https://github.com/skypilot-org/skypilot/pull/1055 * Improve error messages for cloud import errors https://github.com/skypilot-org/skypilot/pull/1156 * Change `~/.ssh/config` permissions https://github.com/skypilot-org/skypilot/pull/1174 * Relative cluster yaml https://github.com/skypilot-org/skypilot/pull/1176 * UX: remove DURATION, move HOURLY_PRICE in status table (-a) https://github.com/skypilot-org/skypilot/pull/1129 ## Thanks to all Contributors! New contributors * @sumanthgenz made their first contribution in https://github.com/skypilot-org/skypilot/pull/1065 * @ewzeng made their first contribution in https://github.com/skypilot-org/skypilot/pull/1174 Many thanks to all contributors who contributed to this release! @Michaelvll, @concretevitamin, @infwinston, @michaelzhiluo, @WoosukKwon, @romilbhardwaj, @sumanthgenz, @ewzeng, @iojw, @franklsf95