MLA-C01 Practice Test 42 — 15 Questions

Question 1

A company needs to deploy a large language model (LLM) on SageMaker with the Triton Inference Server to maximize GPU utilization and reduce latency. They have an NVIDIA A100 GPU. Which SageMaker inference option supports Triton?

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

Accepted Answer

SageMaker real-time endpoint using a Triton Inference Server container. SageMaker real-time endpoints support the Triton Inference Server through a pre-built container that integrates with NVIDIA A100 GPUs, enabling dynamic batching and concurrent model execution to maximize GPU utilization and reduce latency. Triton is designed for high-throughput inference on GPU hardware, making it the correct choice for this scenario.

Answer

SageMaker Batch Transform with Triton

Answer

SageMaker Serverless Inference with a custom container

Answer

SageMaker Neo compiled model on a CPU endpoint

Question 2

A team wants to automatically retrain a model whenever data drift is detected on their SageMaker endpoint. Which AWS service should they use to invoke a retraining pipeline in response to a CloudWatch Alarm?

Accepted Answer

Amazon SNS topic triggering an AWS Lambda function. Amazon SNS can directly subscribe to a CloudWatch Alarm and, upon alarm state change, publish a message to an SNS topic. That topic can then trigger an AWS Lambda function, which invokes the retraining pipeline (e.g., SageMaker Processing or training job). This creates a fully managed, serverless event-driven workflow without needing custom polling or additional orchestration services.

Answer

AWS Step Functions directly from CloudWatch Alarm

Answer

SageMaker Processing job scheduled via EventBridge

Answer

Amazon SQS queue polling by a custom application

Question 3

An ML engineer is using Amazon SageMaker Automatic Model Tuning (AMT) to optimize hyperparameters for a gradient boosting model. The tuning job is taking a long time and has completed many training jobs. The engineer wants to stop training jobs that are unlikely to improve the objective metric. What should they configure?

Accepted Answer

Enable early stopping in the hyperparameter tuning job. Enabling early stopping in the Amazon SageMaker Automatic Model Tuning (AMT) job allows the tuning job to automatically stop training jobs that are unlikely to improve the objective metric based on intermediate results. This reduces the total time and compute cost by terminating poorly performing trials early, which directly addresses the engineer's goal of stopping unpromising training jobs.

Answer

Reduce the number of hyperparameter ranges

Answer

Use a random search strategy instead of Bayesian

Answer

Increase the maximum number of training jobs

Question 4

A machine learning engineer is evaluating a binary classification model that predicts customer churn. The model achieves 95% accuracy, but the engineer suspects class imbalance is causing a misleading metric. Which THREE evaluation steps should the engineer perform to properly assess the model? (Choose THREE.)

Accepted Answer

Calculate precision, recall, and F1-score. Accuracy is misleading for imbalanced datasets. Confusion matrix, precision/recall/F1, and AUC-ROC are robust metrics. RMSE is for regression. MAE is also for regression.

Answer

Calculate RMSE

Answer

Compute Mean Absolute Error (MAE)

Question 5

A company trains a model daily using Amazon SageMaker and uses the model for real-time inference. They want to detect data drift between the training data and the inference data to decide when to retrain. Which AWS service should they use for this purpose?

Accepted Answer

Amazon SageMaker Model Monitor. Amazon SageMaker Model Monitor is the correct service because it is specifically designed to continuously monitor machine learning models in production for data drift, feature attribution drift, and quality issues. It compares the distribution of live inference data against the baseline training data statistics and alerts when drift exceeds defined thresholds, enabling timely retraining decisions.

Answer

Amazon Athena

Answer

AWS Glue

Answer

AWS Lambda

Question 6

A company deploys a model for credit risk assessment on a SageMaker endpoint. To comply with internal policies, they must ensure that the endpoint only allows inference requests from within a specific VPC and that the data is encrypted at rest. Which configuration meets these requirements?

Accepted Answer

Configure endpoint in VPC-only mode and enable KMS encryption for the endpoint. Configuring the SageMaker endpoint in VPC-only mode ensures that inference requests can only originate from within the specified VPC, satisfying the network restriction requirement. Enabling KMS encryption for the endpoint encrypts the model data and inference data at rest using a customer-managed key, meeting the encryption-at-rest policy.

Answer

Enable network isolation mode and use KMS for endpoint encryption

Answer

Configure endpoint in VPC-only mode and enable inter-container traffic encryption

Answer

Deploy endpoint in a public subnet and use a security group to restrict traffic

Question 7

A team is deploying a model that requires GPU acceleration for inference. They are using an Amazon SageMaker real-time endpoint. The model is a large language model (LLM) that does not fit on a single GPU. Which configuration should they use to minimize latency while fitting the model?

Accepted Answer

Use SageMaker's model parallelism library to shard the model across multiple GPUs in a single instance.. SageMaker's model parallelism library allows you to shard a large language model across multiple GPUs within a single instance, enabling inference for models that exceed a single GPU's memory. This approach minimizes latency by keeping all GPUs in a single instance with high-speed interconnects (e.g., NVLink), avoiding the network overhead of distributing across separate instances.

Answer

Use data parallelism with Horovod to distribute inference across GPUs.

Answer

Optimize the model with SageMaker Neo to reduce its size.

Answer

Deploy the model across multiple endpoints and use a load balancer.

Question 8

A company is using SageMaker to train a neural network for image classification. The training job is taking too long. The team wants to reduce training time without sacrificing model accuracy. Which approach should they recommend?

Accepted Answer

Use a GPU-based instance such as ml.p3.2xlarge. GPU-based instances like ml.p3.2xlarge are specifically designed for parallel processing of matrix operations, which are fundamental to neural network training. By offloading compute-intensive tensor operations to GPU cores, training time can be significantly reduced without altering the model architecture or data, thus preserving accuracy.

Answer

Increase the batch size to the maximum possible

Answer

Use a learning rate scheduler that reduces the learning rate over time

Answer

Add more convolutional layers to the model

Question 9

A team wants to apply a custom container for inference on SageMaker. The container needs to implement a web server that responds to API requests. Which protocol and port must the container listen on to be compatible with SageMaker hosting?

Accepted Answer

The container must listen on port 8080 and use HTTP protocol.. SageMaker requires custom inference containers to listen on port 8080 and communicate over HTTP (not HTTPS). The SageMaker hosting service uses a proxy that terminates HTTPS and forwards plain HTTP requests to the container on port 8080. This ensures compatibility with the built-in model serving infrastructure.

Answer

The container must listen on port 8080 and use HTTPS protocol.

Answer

The container can listen on any port as long as the port is specified in the endpoint configuration.

Answer

The container must listen on port 8000 and use HTTP protocol.

Question 10

Which SageMaker feature compiles a trained model into an optimized binary for a specific hardware target (e.g., Intel, ARM, NVIDIA, or edge devices) to improve inference performance?

Accepted Answer

SageMaker Neo. SageMaker Neo is a model compilation service that optimizes models for specific hardware targets. Amazon Elastic Inference attaches GPU acceleration to endpoints, but does not compile models. Model Monitor monitors quality. SageMaker Clarify explains predictions.

Answer

SageMaker Model Monitor

Answer

Amazon Elastic Inference

Answer

SageMaker Clarify

Question 11

A data science team has trained a PyTorch model for real-time inference and needs to deploy it on AWS with GPU acceleration while minimizing cold-start latency. Which SageMaker inference option should they choose?

Accepted Answer

Real-time endpoint with ml.g4dn instance. Real-time endpoints with GPU instances (e.g., ml.g4dn) provide low latency and support GPU acceleration, suitable for interactive inference. Serverless inference does not support GPU instances, asynchronous inference is for non-real-time, and batch transform is for offline predictions.

Answer

Serverless inference

Answer

Batch transform

Answer

Asynchronous inference endpoint

Question 12

A company wants to use SageMaker to deploy a model that requires GPU acceleration for inference but also needs to keep costs low when traffic is low. Which SageMaker feature should they use?

Accepted Answer

SageMaker Elastic Inference. SageMaker Elastic Inference (EI) allows you to attach a fraction of a GPU to a SageMaker endpoint for inference, providing GPU acceleration at a lower cost than using a full GPU instance. This is ideal for scenarios with variable traffic because you can scale the EI accelerator independently of the instance, and pay only for the accelerator when it's used, keeping costs low during low-traffic periods.

Answer

SageMaker Debugger

Answer

SageMaker Managed Spot Training

Answer

SageMaker Model Monitor

Question 13

A company needs to give a data science team in another AWS account access to deploy a model from a shared model registry. Which approach should they use to grant cross-account access?

Accepted Answer

Attach a resource-based policy to the model registry granting the other account access. Resource-based policies (also called resource policies) can be attached to the model registry to allow cross-account access. IAM roles are used for intra-account access, not for granting access to resources in another account.

Answer

Create an IAM role in the other account with permissions to access the registry

Answer

Use AWS Organizations SCP to allow cross-account access

Answer

Share the model artifacts via Amazon S3 bucket policy and use an IAM role

Question 14

A company wants to automatically trigger model retraining when SageMaker Model Monitor detects data drift. Which TWO services should they integrate to achieve this automation? (Choose two.)

Accepted Answer

AWS Lambda. EventBridge can schedule or trigger events based on monitoring job completions, and Lambda can execute the retraining code. SNS is used for notifications but not direct invocation of retraining. Step Functions can orchestrate but is not required for this simple trigger.

Answer

AWS Step Functions

Answer

Amazon CloudWatch Alarms

Answer

Amazon SNS

Question 15

A machine learning team uses SageMaker Pipelines and wants to automatically retrain a model when data drift is detected. They have set up Model Monitor to publish drift violations to CloudWatch. Which approach provides a COMPLETE serverless retraining pipeline triggered by drift detection?

Accepted Answer

Configure a CloudWatch Alarm on drift metric → SNS topic → Lambda function that starts the SageMaker Pipeline execution. The recommended pattern: CloudWatch Alarm triggers on drift metric → SNS message → Lambda function (receives SNS) → starts SageMaker Pipeline execution. EventBridge could also trigger on SNS events, but Lambda is simplest. EventBridge can schedule retraining but does not directly react to specific drift alarms. Step Functions would add unnecessary complexity.

Answer

Use SageMaker Model Monitor to directly invoke a SageMaker Pipeline when drift is detected

Answer

Use EventBridge to schedule retraining daily regardless of drift

Answer

Create an EventBridge rule that triggers on Model Monitor drift events to start the pipeline