nerdexam
Amazon

MLS-C01 · Question #330

An online delivery company wants to choose the fastest courier for each delivery at the moment an order is placed. The company wants to implement this feature for existing users and new users of its…

The correct answer is B. Prepare an Amazon SageMaker Docker container based on the open-source multi-model server. To host multiple XGBoost models efficiently with minimal operational overhead and low instance utilization, using Amazon SageMaker's multi-model endpoint feature with an open-source multi-model server is the most suitable solution.

Machine Learning Implementation and Operations

Question

An online delivery company wants to choose the fastest courier for each delivery at the moment an order is placed. The company wants to implement this feature for existing users and new users of its application. Data scientists have trained separate models with XGBoost for this purpose, and the models are stored in Amazon S3. There is one model for each city where the company operates. Operation engineers are hosting these models in Amazon EC2 for responding to the web client requests, with one instance for each model, but the instances have only a 5% utilization in CPU and memory. The operation engineers want to avoid managing unnecessary resources. Which solution will enable the company to achieve its goal with the LEAST operational overhead?

Options

  • ACreate an Amazon SageMaker notebook instance for pulling all the models from Amazon S3
  • BPrepare an Amazon SageMaker Docker container based on the open-source multi-model server.
  • CKeep only a single EC2 instance for hosting all the models. Install a model server in the instance
  • DPrepare a Docker container based on the prebuilt images in Amazon SageMaker. Replace the

How the community answered

(36 responses)
  • A
    14% (5)
  • B
    78% (28)
  • C
    3% (1)
  • D
    6% (2)

Why each option

To host multiple XGBoost models efficiently with minimal operational overhead and low instance utilization, using Amazon SageMaker's multi-model endpoint feature with an open-source multi-model server is the most suitable solution.

ACreate an Amazon SageMaker notebook instance for pulling all the models from Amazon S3

Amazon SageMaker notebook instances are for development and experimentation, not for hosting production-grade inference endpoints to respond to web client requests.

BPrepare an Amazon SageMaker Docker container based on the open-source multi-model server.Correct

Preparing an Amazon SageMaker Docker container based on the open-source multi-model server allows hosting multiple models on a single SageMaker endpoint, dynamically loading and unloading them as needed, which significantly reduces operational overhead and addresses the low utilization of individual instances.

CKeep only a single EC2 instance for hosting all the models. Install a model server in the instance

While hosting all models on a single EC2 instance reduces the number of instances, it still involves manual management of the EC2 instance and model server, incurring more operational overhead compared to a fully managed SageMaker multi-model endpoint.

DPrepare a Docker container based on the prebuilt images in Amazon SageMaker. Replace the

Using a generic SageMaker prebuilt image for a Docker container typically hosts a single model or requires custom logic for multiple models, lacking the optimized, dynamic model loading capabilities and reduced operational overhead of a dedicated multi-model server.

Concept tested: Amazon SageMaker multi-model endpoints

Source: https://docs.aws.amazon.com/sagemaker/latest/dg/multi-model-endpoints.html

Topics

#Model Deployment#MLOps#SageMaker#Multi-Model Serving

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full MLS-C01 Practice