Zeema
New design
Catalogue / big-data-cluster

Cluster Director

gcpBig data clustergoogle-cluster-director

Managed AI and HPC cluster of accelerator VMs with Slurm scheduling and autohealing.

Provider
gcp
Kind
Big data cluster
Category
Compute
Official docs
https://cloud.google.com/cluster-director/docs/overview
Properties · 4 set by this component
propertytypevaluesrequiredkind defaultthis component
autohealing · component propertyboolean—notrueinherited
autoscalingboolean—nofalseinherited
engineenumspark | hadoop | presto | flink | hpc | ray | dask | kubernetes-batchno—▸ hpc
managedboolean—nofalse▸ true
orchestrator · component propertyenumslurmnoslurminherited
runtimeVersionstring—no—inherited
spotInstancesboolean—nofalse▸ false
topologyAwareScheduling · component propertyboolean—notrueinherited
workerNodesinteger—no—inherited
zoneRedundantboolean—no—▸ false
Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

Stateful node without zone redundancyerror
stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

  • — all of these hold:
  • — the node's `stateless` is false
  • — the node sits inside exactly 1 availability-zone group
  • — this does not hold:
  • — the node sits inside at least one cell group
Use it in a design
nodes:
  - { id: "cluster-director", component: "google-cluster-director" }
Other Big data cluster products · 25
Elastic High Performance Computingalibaba-e-hpcalibabacloudbig-data-clusterE-MapReducealibaba-e-mapreducealibabacloudbig-data-clusterAWS Clean Roomsaws-clean-roomsawsbig-data-clusterAmazon EMRaws-emrawsbig-data-clusterAmazon EMR Serverlessaws-emr-serverlessawsbig-data-clusterAWS Parallel Computing Serviceaws-pcsawsbig-data-clusterAmazon SageMaker HyperPodaws-sagemaker-hyperpodawsbig-data-clusterAzure CycleCloudazure-cyclecloudazurebig-data-clusterAzure Databricksazure-databricksazurebig-data-clusterAzure HDInsightazure-hdinsightazurebig-data-clusterArmadaarmadabig-data-clusterGeneric big data clusterbig-data-clusterbig-data-clusterDaskdaskbig-data-clusterDatabricksdatabricksbig-data-clusterApache Hadoop YARNhadoop-yarnbig-data-clusterRayraybig-data-clusterSlurmslurmbig-data-clusterApache Spark clusterspark-clusterbig-data-clusterVolcanovolcanobig-data-clusterDataprocgoogle-dataprocgcpbig-data-clusterManaged Service for Apache Spark (serverless)google-dataproc-serverlessgcpbig-data-clusterIBM Analytics Engineibm-analytics-engineibmbig-data-clusterOCI AI Data Platformoci-ai-data-platformocibig-data-clusterOCI Big Data Serviceoci-big-data-serviceocibig-data-clusterOCI Data Flowoci-data-flowocibig-data-cluster
Source
# yaml-language-server: $schema=../../schema/component.schema.json
# Cluster Director: automated lifecycle management for AI and HPC clusters of compute- and
# accelerator-optimized Compute Engine instances ("Cluster Director overview"). It ships a managed
# Slurm environment for job scheduling, topology-aware scheduling that colocates instances, and
# autohealing that replaces failed instances; capacity is scaled by editing machine configurations.
schemaVersion: "1"
type: "component"
id: "google-cluster-director"
kind: "big-data-cluster"
provider: "gcp"
docs: "https://cloud.google.com/cluster-director/docs/overview"
defaults:
  managed: true
  engine: "hpc"
  spotInstances: false
  zoneRedundant: false
properties:
  orchestrator:
    type: "enum"
    values: ["slurm"]
    default: "slurm"
  autohealing:
    type: "boolean"
    default: true
  topologyAwareScheduling:
    type: "boolean"
    default: true

The same page as markdown, for an agent: /catalogue/components/google-cluster-director.md