I would like to get the following information using the API.
- Jobs that has been run and the price for running them
- Currently running jobs and their current price
- Total price on each queue from first run
Need to include IB drivers and Nvidia driver in GPU VM sku's while they spin up from schdulers in future cyclecloud releases.1 vote
Accelerated networking disturbs the InfiniBand configuration with the CentOS-HPC images.1 vote
Setting up a job for the first time often requires some trial and error. The autoscale feature in this case is rather contra productive. A cluster can shut down at the wrong time and set one back 15min. When the job is finally running fine, I want to be able to turn on autoscale and walk away.
This is probably scheduler dependent. I suggest to add this feature to Slurm.1 vote
Badaboom BB is arranged by Blambot and it is a not too bad Comic text style that is open for nothing download for your own utilization. furthermore, Badaboom Font free structure contacts the capitalized and lowercase letter with some unique character.
You mention that cyclecloud can create reports of cluster usage per job/user/group/etc but it does not work with PBSPRO scheduler. What is the roadmap for Reports feature inside cyclecloud? Thanks1 vote
- Don't see your idea?