Utilizzare CreateJobQueue con un SDK AWS o una CLI

Come creare una coda di processi a bassa priorità con un unico ambiente di calcolo

Questo esempio crea una coda di processi denominata LowPriority che utilizza l’ambiente di calcolo M4Spot.

Comando:

aws batch create-job-queue --cli-input-json file://<path_to_json_file>/LowPriority.json

Formato del file JSON:

{
  "jobQueueName": "LowPriority",
  "state": "ENABLED",
  "priority": 10,
  "computeEnvironmentOrder": [
    {
      "order": 1,
      "computeEnvironment": "M4Spot"
    }
  ]
}

Output:

{
    "jobQueueArn": "arn:aws:batch:us-east-1:012345678910:job-queue/LowPriority",
    "jobQueueName": "LowPriority"
}

Come creare una coda di processi ad alta priorità con due ambienti di calcolo

Questo esempio crea una coda di processi denominata HighPriority che utilizza l’ambiente di calcolo C4OnDemand con ordine pari a 1 e l’ambiente di calcolo M4Spot con ordine pari a 2. Il pianificatore tenterà innanzitutto di inserire i processi nell’ambiente di calcolo C4OnDemand.

Comando:

aws batch create-job-queue --cli-input-json file://<path_to_json_file>/HighPriority.json

Formato del file JSON:

{
  "jobQueueName": "HighPriority",
  "state": "ENABLED",
  "priority": 1,
  "computeEnvironmentOrder": [
    {
      "order": 1,
      "computeEnvironment": "C4OnDemand"
    },
    {
      "order": 2,
      "computeEnvironment": "M4Spot"
    }
  ]
}

Output:

{
    "jobQueueArn": "arn:aws:batch:us-east-1:012345678910:job-queue/HighPriority",
    "jobQueueName": "HighPriority"
}

    /**
     * Creates a job queue asynchronously.
     *
     * @param jobQueueName the name of the job queue to create
     * @param computeEnvironmentName the name of the compute environment to associate with the job queue
     * @return a CompletableFuture that completes with the Amazon Resource Name (ARN) of the job queue
     */
    public CompletableFuture<String> createJobQueueAsync(String jobQueueName, String computeEnvironmentName) {
        if (jobQueueName == null || jobQueueName.isEmpty()) {
            throw new IllegalArgumentException("Job queue name cannot be null or empty");
        }
        if (computeEnvironmentName == null || computeEnvironmentName.isEmpty()) {
            throw new IllegalArgumentException("Compute environment name cannot be null or empty");
        }

        CreateJobQueueRequest request = CreateJobQueueRequest.builder()
            .jobQueueName(jobQueueName)
            .priority(1)
            .computeEnvironmentOrder(ComputeEnvironmentOrder.builder()
                .computeEnvironment(computeEnvironmentName)
                .order(1)
                .build())
            .build();

        CompletableFuture<CreateJobQueueResponse> response = getAsyncClient().createJobQueue(request);
        response.whenComplete((resp, ex) -> {
            if (ex != null) {
                String errorMessage = "Unexpected error occurred: " + ex.getMessage();
                throw new RuntimeException(errorMessage, ex);
            }
        });

        return response.thenApply(CreateJobQueueResponse::jobQueueArn);
    }