MLA-C01 Deployment and Orchestration of ML Workflows Practice Test 3 — 15 Questions

Question 1

A company needs to deploy a new model version to a SageMaker real-time endpoint. They want to route 5% of traffic to the new version initially to monitor for errors before full rollout. Which deployment strategy should they use?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Canary deployment with production variants. A canary deployment with production variants allows you to route a specific percentage of traffic (e.g., 5%) to the new model version by adjusting the `InitialVariantWeight` parameter in the production variant configuration. This enables gradual traffic shifting while monitoring errors, and you can later increase the weight to 100% for full rollout. SageMaker real-time endpoints support this natively by hosting multiple model variants behind the same endpoint.

Answer

Blue/green deployment

Answer

Shadow testing

Answer

Multi-model endpoint

Question 2

An ML engineer needs to compile a trained TensorFlow model to run efficiently on a target edge device with an ARM CPU. Which AWS service should they use?

Accepted Answer

SageMaker Neo. SageMaker Neo compiles trained models for specific hardware targets, including ARM CPUs, to optimize inference performance.

Answer

SageMaker Debugger

Answer

AWS Inferentia

Answer

Amazon Elastic Inference

Question 3

A data science team uses SageMaker Pipelines for automated training. They need to conditionally register a model only if evaluation metrics exceed a threshold. Which pipeline step type should they use after the evaluation step?

Accepted Answer

Condition step. The Condition step evaluates a condition and branches the pipeline; if the condition is met, the pipeline proceeds to register the model.

Answer

Processing step

Answer

Transform step

Answer

RegisterModel step

Question 4

A company wants to serve a large ensemble of models using NVIDIA Triton Inference Server on SageMaker for high throughput GPU inference. Which SageMaker inference option supports this?

Accepted Answer

Real-time endpoint with a custom container running Triton. SageMaker supports Triton Inference Server through a custom real-time endpoint container, as Triton is optimized for GPU serving on NVIDIA hardware.

Answer

Asynchronous Inference

Answer

Multi-model endpoint

Answer

Serverless Inference

Question 5

A company uses SageMaker Model Registry to manage model versions. They want to automate the approval of models that pass automated evaluation, but require manual approval for others. Which Model Registry feature supports this workflow?

Accepted Answer

Approval workflow via pipeline Condition step. Model Registry supports approval statuses (PendingManualApproval, Approved, Rejected). Automated evaluation can set status to Approved, while borderline cases can be set to PendingManualApproval.

Answer

Cross-account deployment

Answer

Model versioning

Answer

Model lineage

Question 6

A company needs to deploy a model that processes large payloads (up to 1 GB) asynchronously. The results should be written to S3, and the team needs SNS notifications upon completion. Which SageMaker inference option is MOST suitable?

Accepted Answer

Asynchronous Inference. Asynchronous Inference is designed for large payloads, processes requests asynchronously, and supports SNS notifications on completion.

Answer

Batch Transform

Answer

Real-time endpoint

Answer

Serverless Inference

Question 7

An ML team uses AWS Step Functions to orchestrate a retraining pipeline triggered by EventBridge when new training data arrives. The pipeline includes a SageMaker training job and a model evaluation. If evaluation fails, the team wants to send an alert. How should they implement this?

Accepted Answer

Add a Catch rule in the Step Functions state machine to invoke a Lambda alert function. Step Functions supports error handling via Catch rules; a Catch on the training or evaluation task can transition to a Lambda function that sends an alert.

Answer

Use SQS dead-letter queue for failed training jobs

Answer

Configure SageMaker training job to publish to SNS on failure

Answer

Use EventBridge to monitor the training job status

Question 8

A startup wants to deploy a containerized ML application that includes both a model inference server and a preprocessing component in the same endpoint. Which SageMaker endpoint type supports running multiple containers?

Accepted Answer

Multi-container endpoint. Multi-container endpoints allow running multiple containers, enabling preprocessing and inference in the same endpoint.

Answer

Asynchronous Inference

Answer

Multi-model endpoint

Answer

Real-time endpoint

Question 9

An organization wants to automate ML retraining using an event-driven architecture. Which THREE services should they combine? (Select THREE.)

Accepted Answer

SageMaker (training jobs or pipelines). Amazon SageMaker provides the training jobs and pipelines that execute the ML retraining workflow. Amazon EventBridge acts as the event bus that triggers retraining based on events such as new data arrival or model drift detection. AWS Lambda serves as the lightweight compute layer that can preprocess events, invoke SageMaker APIs, or orchestrate conditional logic before starting a training job.

Answer

AWS Glue

Answer

Amazon CloudWatch Logs

Question 10

A team uses SageMaker Pipelines to train and evaluate a model. They want to run the training step only if the data quality check passes, otherwise skip. Which TWO pipeline step types are required? (Select TWO.)

Accepted Answer

Condition step. The Condition step (B) is required because SageMaker Pipelines uses a Condition step to evaluate a boolean expression—such as whether a data quality check passed—and then conditionally execute subsequent steps. The Training step (D) is required because it is the step that actually runs the model training job, and it must be placed inside the 'If' branch of the Condition step to run only when the condition is true.

Answer

RegisterModel step

Answer

Processing step

Answer

Transform step

Question 11

A machine learning engineer needs to deploy a model that requires less than 100 ms inference latency for real-time predictions. The model is a small PyTorch model that fits in a single GPU. Which SageMaker inference option is MOST cost-effective for this scenario?

Accepted Answer

Serverless inference with max concurrency set to 10. For low latency and occasional traffic, serverless inference is cost-effective because it scales to zero when not in use and charges per inference. Real-time endpoints incur cost even when idle, batch transform is for offline processing, and asynchronous inference has higher latency.

Answer

Asynchronous inference endpoint

Answer

Real-time endpoint on ml.g4dn.xlarge

Answer

Batch transform job

Question 12

A data science team uses SageMaker Pipelines to orchestrate their ML workflow. They noticed that even when source data hasn't changed, the pipeline re-runs all steps, wasting compute time. What should they enable to avoid redundant runs?

Accepted Answer

Enable pipeline caching by setting the CacheConfig property for each step. SageMaker Pipelines supports step caching via the `CacheConfig` property. When enabled, the pipeline checks if the step's inputs (including source data, parameters, and code) have changed since the last successful run. If no changes are detected, the step is skipped and the previous output is reused, eliminating redundant compute.

Answer

Configure the pipeline to run on a schedule instead of on-demand

Answer

Use the Parameter step to pass previous execution ID

Answer

Use Lambda step to check data changes before running

Question 13

A team uses MLflow on SageMaker for experiment tracking. They want to automatically deploy the best-performing model from an MLflow run to a SageMaker endpoint for real-time inference. What is the MOST efficient way to achieve this?

Accepted Answer

Use SageMaker Pipelines with the MLflow integration to register the model and deploy via a Transform step. The MLflow Model Registry can be integrated with SageMaker via the MLflow plugin for SageMaker, which allows direct deployment from the registry to an endpoint. Alternatively, using SageMaker Pipelines with the MLflow integration is more automated and production-grade.

Answer

Use AWS Step Functions to trigger an MLflow run and then call SageMaker CreateEndpoint

Answer

Set up an EventBridge rule to trigger a Lambda that deploys the model whenever a new MLflow run is logged

Answer

Manually export the model artifact from MLflow and upload to S3, then create a SageMaker model and endpoint

Question 14

A company uses SageMaker Model Registry to manage model versions. They have a cross-account deployment requirement: models approved in the development account must be deployed to a production account. Which approach is the MOST secure and recommended?

Accepted Answer

Share the model package group from the development account to the production account using AWS RAM, then create a model version in the production account. Cross-account deployment can be achieved by sharing the model package across accounts using AWS Resource Access Manager (RAM) or by exporting the model artifact to an S3 bucket with appropriate cross-account permissions, then creating the model in the target account.

Answer

Export the model from Model Registry to a tar.gz file and upload to the production account manually

Answer

Copy the model artifact to a public S3 bucket and then create the model in the production account

Answer

Use a Lambda function in the development account to call CreateEndpoint in the production account using cross-account IAM roles

Question 15

A team wants to use AWS Step Functions to orchestrate a retraining workflow that is triggered when new data arrives in an S3 bucket. They also need to monitor model drift. Which event-driven approach should they use?

Accepted Answer

Configure EventBridge to capture S3 PutObject events and target an AWS Step Functions state machine that runs the retraining pipeline. AWS EventBridge can capture S3 PutObject events (via S3's default event notifications or a more granular EventBridge rule) and directly target a Step Functions state machine as a target. This creates a fully event-driven, serverless orchestration for the retraining pipeline without polling or custom code. Step Functions then coordinates the retraining steps, including model drift monitoring, in a reliable and auditable manner.

Answer

Use a cron-based Step Function schedule that checks for new data every hour

Answer

Set up an S3 event notification to invoke a Lambda function that starts a SageMaker training job directly

Answer

Use SageMaker Pipelines with a schedule trigger