Michael Luo
Michael Luo
After some discussion, the conclusion from last week to align user and admin's python (for submitting Ray jobs, since the Ray cluster is launched via admin's python) is to add...
Refactors local cloud checking methods from `cli.py` into `backends/onprem_utils.py`
## Features - [ ] `sky admin status` - New CLI command that lists all user jobs for the administrator - [ ] Add local cluster spillover to multiple clouds...
Add a Sky Storage Download Feature. Also fixes #932. To download a folder from a bucket using Sky Storage, run: ``` from sky.data import storage, StoreType storage_obj = storage.Storage(name='imagenet-bucket') storage_obj.add_store(StoreType.S3)...
In `head_start_ray_commands` for every Cloud jinja file, there is the command: ``` (ps aux | grep "-m sky.skylet.skylet" | grep -q python3) || nohup python3 -m sky.skylet.skylet >> ~/.sky/skylet.log 2>&1...
Replication: Run `sky launch -c hello storage.yaml` with a filemount, while the Storage object is syncing, run `sky storage delete ...`
1. Launch P3.8x with 4 V100s. 2. Submit job with 2 V100. 3. Ray Status on the head node seems to have incorrect logging. This could be a Ray bug.
## Description Related to #388. When Tianjun tried to set his environment variables and setup his work directory, these factors depended on the work directory being named `nce_spede`. This, however,...
If the user's setup script installs their own `workdir` (such as `pip install -e .`), the installed packaged will be named `sky_workdir` because Sky uploads the user's work directory under...