Sidecar S3 Gateway

You can interact with input/output data through the S3 protocol using HPE Machine Learning Data Management’s S3-protocol-enabled pipelines.


HPE Machine Learning Data Management’s S3-protocol-enabled pipelines run a separate S3 gateway instance in a sidecar container within the pipeline-worker pod. Using this approach enables maintaining data provenance since the external code (e.g., within a Kubeflow pod) is executed in (and associated with) a HPE Machine Learning Data Management job.

When enabled, input and output repositories are exposed as S3 Buckets via the S3 gateway sidecar instance.

  • Input Address: s3://<input_repo_name>.
  • Output Address: s3://out

Example with Kubeflow Pod

The following diagram shows communication between the S3 gateway deployed in a sidecar and the Kubeflow pod.

Kubeflow S3 gateway

Configure an S3-enabled Pipeline

  1. Open your pipeline spec.
  2. Add "s3": true to input.pfs.
  3. Add "s3Out": true to pipeline.
  4. Save your spec.
  5. Update your pipeline.

Example Pipeline Spec

The following spec example reads files in the input bucket labresults and copies them in the pipeline’s output bucket:

  "pipeline": {
    "name": "s3_protocol_enabled_pipeline"
  "input": {
    "pfs": {
      "glob": "/",
      "repo": "labresults",
      "name": "labresults",
      "s3": true
  "transform": {
    "cmd": [ "sh" ],
    "stdin": [ "set -x && mkdir -p /tmp/result && aws --endpoint-url $S3_ENDPOINT s3 ls && aws --endpoint-url $S3_ENDPOINT s3 cp s3://labresults/ /tmp/result/ --recursive && aws --endpoint-url $S3_ENDPOINT s3 cp /tmp/result/ s3://out --recursive" ],
    "image": "pachyderm/ubuntu-with-s3-clients:v0.0.1"
  "s3Out": true

User Code Requirements

Your user code is responsible for:

  • Providing its own S3 client package as part of the image (boto3)
  • reading and writing in the S3 Buckets exposed to the pipeline

Accessing the Sidecar

Use the S3_ENDPOINT environment variable to access the sidecar. No authentication is needed; you can only read the input bucket and write in the output bucket.

aws --endpoint-url $S3_ENDPOINT s3 cp /tmp/result/ s3://out --recursive

Triggering External Pipelines

If Authentication is enabled, you can access the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY env vars in your pipeline user code to forward your pipeline’s auth credentials to third-party tools like Spark.


  • All files are processed as a single datum, meaning:
    • The glob field in the pipeline must be set to "glob": "/".
    • Already processed datums are not skipped.
  • Only cross inputs are supported; join, group, and union are not supported.
  • You can create a cross of an S3-enabled input with a non-S3 input; For a non-S3 input in such a cross, you can still specify a glob pattern.
  • Input bucket(s) are read-only, and the output bucket is initially empty and writable.