Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
81 changes: 0 additions & 81 deletions examples/gke-a3-megagpu/chs-cronjob.yaml.tftpl

This file was deleted.

73 changes: 0 additions & 73 deletions examples/gke-a3-megagpu/chs-permissions.yaml.tftpl

This file was deleted.

26 changes: 0 additions & 26 deletions examples/gke-a3-megagpu/chs-pvc.yaml.tftpl

This file was deleted.

3 changes: 0 additions & 3 deletions examples/gke-a3-megagpu/gke-a3-megagpu-deployment.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,3 @@ vars:
# To target a BLOCK_NAME, the name of the extended reservation
# can be inputted as <reservation-name>/reservationBlocks/<reservation-block-name>
reservation: RESERVATION_NAME

enable_periodic_health_checks: false # Make this true to run CHS (healthchecks)
health_check_schedule: "0 0 * * 0" # Run the health check at 12:00 AM (midnight) every Sunday
31 changes: 0 additions & 31 deletions examples/gke-a3-megagpu/gke-a3-megagpu.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -50,15 +50,6 @@ vars:
accelerator_type: nvidia-h100-mega-80gb
version_prefix: "1.32."

enable_periodic_health_checks: false # Make this true to run CHS (healthchecks)
health_check_schedule: "0 0 * * 0" # Run the health check at 12:00 AM (midnight) every Sunday

permissions_file_staged_path: $(ghpc_stage("./chs-permissions.yaml.tftpl"))
chs_output_bucket_name: chs-result
chs_pvc_claim_name: chs-output-pvc
chs_cronjob_rendered_path: $(ghpc_stage("./chs-cronjob.yaml.tftpl"))
chs_pvc_rendered_path: $(ghpc_stage("./chs-pvc.yaml.tftpl"))

deployment_groups:
- group: primary
modules:
Expand Down Expand Up @@ -149,28 +140,6 @@ deployment_groups:
use: [gke_cluster]
settings:
apply_manifests:
- source: $(vars.permissions_file_staged_path)
enable: $(vars.enable_periodic_health_checks)
template_vars:
project_id: $(vars.project_id)
deployment_name: $(vars.deployment_name)
- source: $(vars.chs_pvc_rendered_path)
enable: $(vars.enable_periodic_health_checks)
template_vars:
pvc_name: $(vars.chs_pvc_claim_name)
access_mode: ReadWriteOnce
capacity: 1Gi
storage_class_name: standard-rwo
- source: $(vars.chs_cronjob_rendered_path)
enable: $(vars.enable_periodic_health_checks)
template_vars:
project_id: $(vars.project_id)
deployment_name: $(vars.deployment_name)
region: $(vars.region)
machine_type: a3-megagpu-8g
gcs_bucket: $(vars.chs_output_bucket_name)
gcs_pvc: $(vars.chs_pvc_claim_name)
cronjob_schedule: $(vars.health_check_schedule)
kueue:
install: true
wait: true
Expand Down
67 changes: 0 additions & 67 deletions examples/gke-a3-megagpu/read-chs-logs-job.yaml

This file was deleted.

81 changes: 0 additions & 81 deletions examples/gke-a3-ultragpu/chs-cronjob.yaml.tftpl

This file was deleted.

Loading
Loading