A large machine learning job spans many nodes and runs most efficiently when it has access to all of the hardware resources on each node ... So for many of our workloads, a single pod occupies the entire node
Hm, why not just use the underlying nodes then, without Kubernetes?
Is the underlying cloud that bad at scheduling, and are they keeping the VMs warm all the time?
What are they gaining for this indirection? Is it to get a common interface across GCP and other clouds?
Bin-packing or fragmentation is not a common problem
Hm, why not just use the underlying nodes then, without Kubernetes?
Is the underlying cloud that bad at scheduling, and are they keeping the VMs warm all the time?
What are they gaining for this indirection? Is it to get a common interface across GCP and other clouds?
Bin-packing or fragmentation is not a common problem
there’s relatively low strain on the scheduler.
That said, strain on the kube-scheduler is spiky