MLS-C01 Machine Learning Implementation and Operations Practice Test 22 — 15 Questions

Question 1

A company is deploying a model to an Amazon SageMaker endpoint for real-time inference. The model requires a GPU for low-latency predictions. Which instance type should be chosen?

Accepted Answer

ml.g4dn.xlarge. SageMaker GPU instances (ml.p3, ml.p4, ml.g4dn, ml.g5) are designed for GPU inference. Option D is correct. Option A is wrong because ml.c5 is CPU only. Option B is wrong because ml.m5 is CPU only. Option C is wrong because ml.r5 is CPU only.

Answer

ml.c5.xlarge

Answer

ml.r5.2xlarge

Answer

ml.m5.large

Question 2

A machine learning engineer is using AWS Step Functions to orchestrate a SageMaker training job followed by a Lambda function for post-processing. The training job completes successfully, but the Lambda function fails with a timeout error. What is the MOST likely cause?

Accepted Answer

The Lambda function execution time exceeds the maximum timeout limit. Lambda has a maximum execution timeout of 15 minutes. If post-processing takes longer, it will timeout. Option A is correct. Option B is wrong because the training job completed successfully. Option C is wrong because Step Functions itself is not the cause. Option D is wrong because the failure is a timeout, not a permission issue.

Answer

The Lambda function's IAM role lacks permissions to access the training output

Answer

The Step Functions state machine has a misconfigured retry policy

Answer

The SageMaker training job output data is too large for Lambda to process

Question 3

A company wants to track and compare metrics from multiple machine learning experiments. Which Amazon SageMaker feature should be used?

Accepted Answer

SageMaker Experiments. SageMaker Experiments helps track and compare metrics. Option B is correct. Option A is wrong because Model Monitor is for monitoring drift. Option C is wrong because Debugger is for debugging training. Option D is wrong because Ground Truth is for labeling.

Answer

SageMaker Ground Truth

Answer

SageMaker Model Monitor

Answer

SageMaker Debugger

Question 4

A data scientist uses SageMaker to train a model and wants to automatically stop the training job if the loss is not improving after a certain number of steps. Which feature should be used?

Accepted Answer

SageMaker Debugger. SageMaker Debugger can monitor loss and trigger actions like stopping the job. Option D is correct. Option A is wrong because automatic tuning is for hyperparameter optimization. Option B is wrong because Experiments is for tracking. Option C is wrong because Ground Truth is for labeling.

Answer

SageMaker Experiments

Answer

SageMaker Automatic Model Tuning

Answer

SageMaker Ground Truth

Question 5

A company is using SageMaker to host a model for real-time inference. They notice that the endpoint's latency increases over time. The model is stateless and the inference code does not log any errors. What is the MOST likely cause?

Accepted Answer

Memory leak in the inference container. Memory leaks cause gradual performance degradation. Option A is correct. Option B is wrong because the model is stateless. Option C is wrong because auto scaling would add instances, not degrade existing ones. Option D is wrong because the model is stateless.

Answer

Gradual increase in request payload size

Answer

Endpoint auto scaling is adding new instances

Answer

Model is accumulating state from previous requests

Question 6

Which TWO AWS services can be used to deploy a machine learning model for serverless inference? (Choose 2.)

Accepted Answer

Amazon SageMaker Serverless Inference. SageMaker Serverless Inference (Option A) and AWS Lambda (Option C) both support serverless inference. Option B (Amazon ECS) requires managing clusters. Option D (Amazon EMR) is for big data. Option E (AWS Batch) is for batch computing.

Answer

Amazon EMR

Answer

Amazon ECS with Fargate

Answer

AWS Batch

Question 7

Which THREE actions should be taken to ensure data security when training a model using Amazon SageMaker with data stored in Amazon S3? (Choose 3.)

Accepted Answer

Use a VPC to isolate the SageMaker training job. Encrypt data at rest (Option A), use VPC (Option B), and use an IAM role with least privilege (Option D). Option C is wrong because SageMaker uses S3, not EBS, for training data. Option E is wrong because SageMaker does not support S3 bucket policies directly; IAM roles are used.

Answer

Apply an S3 bucket policy that denies all access except from the SageMaker service

Answer

Attach an EBS volume for storing training data

Question 8

Which TWO approaches can reduce inference latency on a SageMaker real-time endpoint? (Choose 2.)

Accepted Answer

Attach an Elastic Inference accelerator. Using a GPU instance (Option A) and enabling SageMaker Model Monitor (Option E) are not directly for latency reduction. Actually, correct: Option B (Elastic Inference) and Option D (compiled model with SageMaker Neo) reduce latency. Option A is wrong because GPU does not always reduce latency; it can add overhead. Option C is wrong because larger batch sizes increase latency. Option E is wrong because Model Monitor adds overhead.

Answer

Increase the batch size

Answer

Enable SageMaker Model Monitor

Answer

Use a GPU instance type

Question 9

A financial services company uses Amazon SageMaker to train a fraud detection model. The training data is stored in an S3 bucket encrypted with AWS KMS. The SageMaker training job is configured to use a custom Docker container that reads data from S3 and writes model artifacts back to S3. The training job fails with the error: 'Unable to write model artifact to s3://my-bucket/output/model.tar.gz. Access Denied.' The IAM role used by the training job has the following permissions: s3:GetObject and s3:PutObject on the bucket, and kms:Decrypt on the KMS key. The training job is not using a VPC. What is the MOST likely cause of the failure?

Accepted Answer

The IAM role does not have kms:GenerateDataKey permission on the KMS key. Option B is correct because SageMaker training jobs need kms:GenerateDataKey permission to write encrypted objects. Option A is wrong because the role already has s3:PutObject. Option C is wrong because the job is not in a VPC, so no VPC endpoint issues. Option D is wrong because S3 Batch Operations is irrelevant.

Answer

The S3 bucket is in a different region than the training job

Answer

The S3 bucket requires S3 Batch Operations for writing artifacts

Answer

The IAM role does not have s3:PutObject permission on the output bucket

Question 10

A media company uses SageMaker to deploy a real-time inference endpoint for content recommendation. The model is a PyTorch model that uses GPU. The endpoint is deployed with an ml.p3.2xlarge instance. Over time, the endpoint's latency increases significantly during peak hours. The company has enabled auto scaling based on CPU utilization. However, the latency spikes occur even when CPU utilization is low. The model is stateless and the inference code is efficient. What is the MOST likely cause of the latency spikes?

Accepted Answer

Auto scaling is configured based on CPU utilization, but the bottleneck is GPU utilization. Option A is correct because GPU utilization is the correct metric for GPU-based models. Auto scaling based on CPU utilization does not capture GPU load. Option B is wrong because there is no mention of memory issues. Option C is wrong because the model is stateless. Option D is wrong because the model is already on GPU.

Answer

The model uses stateful processing that accumulates requests

Answer

The inference container has a memory leak that causes gradual slowdown

Answer

The instance type is too small for the model

Question 11

A startup is using SageMaker to train a model using the built-in XGBoost algorithm. The training job runs successfully but the resulting model performs poorly on the test data. The data scientist suspects overfitting. The training data is relatively small (10,000 rows). Which action should be taken to reduce overfitting?

Accepted Answer

Decrease the number of trees (num_round) to 50. Option C is correct because reducing the number of trees (or early stopping) reduces overfitting. Option A is wrong because increasing trees increases overfitting. Option B is wrong because increasing learning rate may cause divergence. Option D is wrong because using a larger instance does not affect overfitting.

Answer

Increase the learning rate to 0.3

Answer

Increase the number of trees (num_round) to 500

Answer

Use a larger instance type

Question 12

A data scientist is training a deep learning model on a large dataset using Amazon SageMaker. The training job is taking too long and the scientist wants to reduce the training time by distributing the workload across multiple GPUs. Which SageMaker feature should be used to achieve this?

Accepted Answer

Use SageMaker's distributed training libraries. Option D is correct because SageMaker's distributed training libraries enable efficient distribution across multiple GPUs. Option A is wrong because Hyperparameter Tuning is for optimizing hyperparameters, not for distributed training. Option B is wrong because Automatic Model Tuning is for hyperparameter optimization. Option C is wrong because Amazon EMR is for big data processing, not for deep learning training.

Answer

Use Amazon EMR to distribute the training

Answer

Use SageMaker Automatic Model Tuning

Answer

Use SageMaker Hyperparameter Tuning

Question 13

A team is deploying a machine learning model to production using Amazon SageMaker. They want to automatically scale the endpoint based on the incoming request volume, and they also need to ensure that the endpoint can handle sudden bursts of traffic without dropping requests. Which scaling policy should they use?

Accepted Answer

Target tracking scaling policy based on the number of invocations. Option B is correct because a target tracking scaling policy with a specified target value for the metric allows the endpoint to automatically adjust capacity to maintain the target metric, and it can handle bursts by adding more instances proactively. Option A is wrong because a simple scaling policy based on average latency may not handle bursts quickly. Option C is wrong because a scheduled scaling policy is for predictable traffic patterns. Option D is wrong because manual scaling is not automatic.

Answer

Scheduled scaling policy for peak hours

Answer

Simple scaling policy based on average latency

Answer

Manual scaling by monitoring CloudWatch alarms

Question 14

A company has a real-time inference endpoint on Amazon SageMaker that uses a custom container. The endpoint is experiencing high latency and occasional 502 errors. The logs from the container show that the model inference time is low, but the overall response time is high. Which step is MOST likely to reduce the latency?

Accepted Answer

Increase the number of instances and enable auto-scaling. Option C is correct because increasing the endpoint's instance count and enabling auto-scaling can distribute the load and reduce queuing delays. Option A is wrong because the inference time is already low, so optimizing the model further won't help much. Option B is wrong because increasing instance size may help but is less cost-effective than scaling out. Option D is wrong because switching to batch transform is for offline inference, not real-time.

Answer

Switch to batch transform to process requests in batches

Answer

Use a larger instance type for the endpoint

Answer

Optimize the model to reduce inference time

Question 15

A machine learning engineer is deploying a model that was trained on a large dataset stored in Amazon S3. The model needs to be retrained daily with new data. Which approach is the MOST cost-effective for storing the training data while allowing quick access for retraining?

Accepted Answer

Use S3 Intelligent-Tiering. Option B is correct because S3 Intelligent-Tiering automatically moves data between access tiers to optimize costs, and it provides low-latency access. Option A is wrong because S3 Standard is more expensive for data that is not frequently accessed. Option C is wrong because S3 Glacier is for long-term archival, not for data that needs to be accessed daily. Option D is wrong because S3 One Zone-IA is cheaper but may not be suitable for critical data and still costs more than Intelligent-Tiering for varying access patterns.

Answer

Store all data in S3 Standard

Answer

Use S3 Glacier Deep Archive

Answer

Use S3 One Zone-IA