Description

The AWS Lambda batch materialization engine is considered alpha status. It relies on the offline store to output feature values to S3 via to_remote_storage, and then loads them into the online store.

See LambdaMaterializationEngineConfig for configuration options.

See also Dockerfile for a Dockerfile that can be used below with materialization_image.

Example

...
offline_store:
  type: snowflake.offline
...
batch_engine:
  type: lambda
  lambda_role: [your iam role]
  materialization_image: [image uri of above Docker image]

Description

The Bytewax batch materialization engine provides an execution engine for batch materializing operations (materialize and materialize-incremental).

Guide

In order to use the Bytewax materialization engine, you will need a Kubernetes cluster running version 1.22.10 or greater.

Kubernetes Authentication

The Bytewax materialization engine loads authentication and cluster information from the kubeconfig file. By default, kubectl looks for a file named config in the $HOME/.kube directory. You can specify other kubeconfig files by setting the KUBECONFIG environment variable.

Resource Authentication

Bytewax jobs can be configured to access Kubernetes secrets as environment variables to access online and offline stores during job runs.

To configure secrets, first create them using kubectl:

kubectl create secret generic -n bytewax aws-credentials --from-literal=aws-access-key-id='<access key id>' --from-literal=aws-secret-access-key='<secret access key>'

If your Docker registry requires authentication to store/pull containers, you can use this same approach to store your repository access credential and use when running the materialization engine.

Then configure them in the batch_engine section of feature_store.yaml:

batch_engine:
  type: bytewax
  namespace: bytewax
  env:
    - name: AWS_ACCESS_KEY_ID
      valueFrom:
        secretKeyRef:
          name: aws-credentials
          key: aws-access-key-id
    - name: AWS_SECRET_ACCESS_KEY
      valueFrom:
        secretKeyRef:
          name: aws-credentials
          key: aws-secret-access-key
  image_pull_secrets:
    - docker-repository-access-secret

Configuration

The Bytewax materialization engine is configured through the The feature_store.yaml configuration file:

batch_engine:
  type: bytewax
  namespace: bytewax
  image: bytewax/bytewax-feast:latest
  image_pull_secrets:
    - my_container_secret
  service_account_name: my-k8s-service-account
  annotations:
    # example annotation you might include if running on AWS EKS
    iam.amazonaws.com/role: arn:aws:iam::<account number>:role/MyBytewaxPlatformRole
  resources:
    limits:
      cpu: 1000m
      memory: 2048Mi
    requests:
      cpu: 500m
      memory: 1024Mi

Notes:

• The namespace configuration directive specifies which Kubernetes namespace jobs, services and configuration maps will be created in.

• The image_pull_secrets configuration directive specifies the pre-configured secret to use when pulling the image container from your registry

• The service_account_name specifies which Kubernetes service account to run the job under

• annotations allows you to include additional Kubernetes annotations to the job. This is particularly useful for IAM roles which grant the running pod access to cloud platform resources (for example).

• The resources configuration directive sets the standard Kubernetes resource requests for the job containers to utilise when materializing data.

Building a custom Bytewax Docker image

The image configuration directive specifies which container image to use when running the materialization job. To create a custom image based on this container, run the following command:

DOCKER_BUILDKIT=1 docker build . -f ./sdk/python/feast/infra/materialization/contrib/bytewax/Dockerfile -t <image tag>

Once that image is built and pushed to a registry, it can be specified as a part of the batch engine configuration:

batch_engine:
  type: bytewax
  namespace: bytewax
  image: <image tag>

Description

The Spark batch materialization engine is considered alpha status. It relies on the offline store to output feature values to S3 via to_remote_storage, and then loads them into the online store.

See SparkMaterializationEngine for configuration options.

Example

...
offline_store:
  type: snowflake.offline
...
batch_engine:
  type: spark.engine
  partitions: [optional num partitions to use to write to online store]

Example in Python

from feast import FeatureStore, RepoConfig
from feast.repo_config import RegistryConfig
from feast.infra.online_stores.dynamodb import DynamoDBOnlineStoreConfig
from feast.infra.offline_stores.contrib.spark_offline_store.spark import SparkOfflineStoreConfig

repo_config = RepoConfig(
    registry="s3://[YOUR_BUCKET]/feast-registry.db",
    project="feast_repo",
    provider="aws",
    offline_store=SparkOfflineStoreConfig(
      spark_conf={
        "spark.ui.enabled": "false",
        "spark.eventLog.enabled": "false",
        "spark.sql.catalogImplementation": "hive",
        "spark.sql.parser.quotedRegexColumnNames": "true",
        "spark.sql.session.timeZone": "UTC"
      }
    ),
    batch_engine={
      "type": "spark.engine",
      "partitions": 10
    },
    online_store=DynamoDBOnlineStoreConfig(region="us-west-1"),
    entity_key_serialization_version=2
)

store = FeatureStore(config=repo_config)

Please see Batch Materialization Engine for an explanation of batch materialization engines.

Description

The batch materialization engine provides a highly scalable and parallel execution engine using a Snowflake Warehouse for batch materializations operations (materialize and materialize-incremental) when using a SnowflakeSource.

The engine requires no additional configuration other than for you to supply Snowflake's standard login and context details. The engine leverages custom (automatically deployed for you) Python UDFs to do the proper serialization of your offline store data to your online serving tables.

When using all three options together, snowflake.offline, snowflake.engine, and snowflake.online, you get the most unique experience of unlimited scale and performance + governance and data security.

Example