smartdispatch
smartdispatch copied to clipboard
SMART-Lab
An easy to use job launcher for supercomputers with PBS compatible job manager.
Requires #166 to be merged - [ ] Add config file ([ref](https://github.com/SMART-Lab/smartdispatch/tree/4c38f0afa3066961a999402ae076744adee124e3/smartdispatch/config)) - [ ] Add cluster name support to detect_cluster() ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L112-L139)) - [ ] Add launcher ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L141-L145)) - [...
Requires #166 to be merged - [ ] Add config file ([ref](https://github.com/SMART-Lab/smartdispatch/tree/4c38f0afa3066961a999402ae076744adee124e3/smartdispatch/config)) - [ ] Add cluster name support to detect_cluster() ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L112-L139)) - [ ] Add launcher ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L141-L145)) - [...
Requires #166 to be merged - [ ] Add config file ([ref](https://github.com/SMART-Lab/smartdispatch/tree/4c38f0afa3066961a999402ae076744adee124e3/smartdispatch/config)) - [ ] Add cluster name support to detect_cluster() ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L112-L139)) - [ ] Add launcher ([ref](https://github.com/aalitaiga/smartdispatch/blob/559de857fe9723ec705a41c054669fa25c1aca99/smartdispatch/utils.py#L141-L145)) - [...
Keeping track of a hyperparameter search is much easier when using smart-dispatch. It would be nice to have the ability to launch jobs with the [] list notation locally without...
As suggested in #112 we should write some documentation. - [ ] Set up a [ReadTheDoc](https://readthedocs.org/), learn the good practices and how to sync it with the README.md - [...
Right now, our config files (JSON) describe the information of available queues on a given cluster and are not easily customizable. For instance, when loading queues from a config file,...
When a job is terminated and needs to be relaunched, aka autoresumed, we should add a dependency on the new job to the current one to prevent some unforeseeable crashes....
It would be nice to have a way of using smartdispatch as a front end for slurm clusters.
I get this error on a cluster with the IBM LSF scheduler (http://wiki.umassrc.org/wiki/index.php/Main_Page) ``` $ smart-dispatch WARNING:root:Cluster does not support flock! Falling back to folder lock. qstat: illegal option --...