xpk
xpk copied to clipboard
google
xpk (Accelerated Processing Kit, pronounced x-p-k,) is a software tool to help Cloud developers to orchestrate training jobs on accelerators such as TPUs and GPUs on GKE.
## Fixes / Features Create the `xpk` package. Created the `utils` module, which now contains utils function from `xpk.py` ## Testing / Documentation Tested manually, installing xpk and running it....
## Fixes / Features - - ## Testing / Documentation Testing details. - [ y/n ] Tests pass - [ y/n ] Appropriate changes to documentation are included in the...
I wonder if `xpk` supports a cluster creation from several reservations. If not, do you have any plans to add this feature?
…eation / deletion ## Fixes / Features - - ## Testing / Documentation Testing details. - [ y/n ] Tests pass - [ y/n ] Appropriate changes to documentation are...
We have a CPU-only cluster -- n2-standard-32-1024 that has 1024 of n2-standard-32 nodes. There, we technically should have a rough 1024 * 32 CPU resources but I'm seeing 1024 nominated...
If there is an upgrad available for the cluster, additional text is printed to stderr ``` * - There is an upgrade available for your cluster(s). To upgrade nodes to...
## Fixes / Features Add xpk info command. There are two subcommands: xpk info localqueues, xpk info clusterqueues, for both of them there is --cluster flag supported. ## Testing /...
## Fixes / Features - Creates Storage CRD that allows attaching existing GCS Bucket to workloads ## Testing / Documentation Tested manually, CI/CD tests in progress - [ y ]...
kueue supports all-or-nothing scheduling: https://kueue.sigs.k8s.io/docs/tasks/manage/setup_wait_for_pods_ready/ Large multi-pod workloads that need every pod to be running to make progress (e.g. single-program-multi-data workloads) can deadlock capacity if the physical availability of resources...
## Fixes / Features - Fixes workload rendering when using spot, without this change xpk workload create errors like: ``` [XPK] Waiting for `Creating Workload`, for 0 seconds error: error...