Process data in an Amazon S3 bucket with Distributed Map

This sample project demonstrates how you can use the Distributed Map state to process large-scale data, for example, analyze historical weather data and identify the weather station that has the highest average temperature on the planet each month. The weather data is recorded in over 12,000 CSV files, which in turn are stored in an Amazon S3 bucket.

This sample project includes two Distributed Map states named Distributed S3 copy NOA Data and ProcessNOAAData. Distributed S3 copy NOA Data iterates over the CSV files in a public Amazon S3 bucket named noaa-gsod-pds and copies them to an Amazon S3 bucket in your AWS account. ProcessNOAAData iterates over the copied files and includes a Lambda function that performs the temperature analysis.

The sample project first checks the contents of the Amazon S3 bucket with a call to the ListObjectsV2 API action. Based on the number of keys returned in response to this call, the sample project takes one of the following decisions:

• If the key count is more than or equal to 1, the project transitions to the ProcessNOAAData state. This Distributed Map state includes a Lambda function named TemperatureFunction that finds the weather station that had the highest average temperature each month. This function returns a dictionary with year-month as the key and a dictionary that contains information about the weather station as the value.

• If the returned key count doesn't exceed 1, the Distributed S3 copy NOA Data state lists all objects from the public bucket noaa-gsod-pds and iteratively copies the individual objects to another bucket in your account in batches of 100. An Inline Map performs the iterative copying of the objects.

  After all objects are copied, the project transitions to the ProcessNOAAData state for processing the weather data.

The sample project finally transitions to a reducer Lambda function that performs a final aggregation of the results returned by the TemperatureFunction function and writes the results to an Amazon DynamoDB table.

With Distributed Map, you can run up to 10,000 parallel child workflow executions at a time. In this sample project, the maximum concurrency of ProcessNOAAData Distributed Map is set at 3000 that limits it to 3000 parallel child workflow executions.

This sample project creates the state machine, the supporting AWS resources, and configures the related IAM permissions. Explore this sample project to learn about using the Distributed Map for orchestrating large-scale, parallel workloads, or use it as a starting point for your own projects.

Important

This sample project is only available in the US East (N. Virginia) Region.

AWS CloudFormation template and additional resources

You use a CloudFormation template to deploy this sample project. This template creates the following resources in your AWS account:

• A Step Functions state machine.

• Execution role for the state machine. This role grants the permissions that your state machine needs to access other AWS services and resources such as the Lambda function's Invoke action.

• An Amazon S3 bucket named NOAADataBucket. This bucket contains the CSV files with weather data.

• A Lambda function named ReducerFunction that performs a final aggregation of the weather data and writes the results to an Amazon DynamoDB table.

• Execution role for the reducer Lambda function. This role grants the function permission to access other AWS services.

• An Amazon S3 output bucket to store the weather analysis results.

• A DynamoDB table named ResultsDynamoDBTable that contains the results returned by the ReducerFunction.

• A Lambda function named TemperatureFunction that finds the highest monthly average temperature.

• Execution role for the Lambda function. This role grants the function permission to access other AWS services.

• A CloudWatch log group that stores information related to the state machine’s execution history.

Important

Standard charges apply for each service.

Step 1: Create the state machine and provision resources

1. Open the Step Functions console and choose Create state machine.

2. Type Distributed Map to process files in S3 in the search box, and then choose Distributed Map to process files in S3 from the search results that are returned.

3. Choose Next to continue.

4. Choose Run a demo to create a read-only and ready-to-deploy workflow, or choose Build on it to create an editable state machine definition that you can build on and later deploy.

   For information about the resources that will be created for this sample project, see AWS CloudFormation template and additional resources.

   The following image shows the workflow graph for the Distributed Map to process files in S3 sample project:

   

5. Choose Use template to continue with your selection.

Next steps depend on your previous choice:

1. Run a demo – You can review the state machine before you create a read-only project with resources deployed by AWS CloudFormation to your AWS account.

   You can view the state machine definition, and when you are ready, choose Deploy and run to deploy the project and create the resources.

   Deploying can take up to 10 minutes to create resources and permissions. You can use the Stack ID link to monitor progress in AWS CloudFormation.

   After deploy completes, you should see your new state machine in the console.

2. Build on it – You can review and edit the workflow definition. You might need to set values for placeholders in the sample project before attemping to run your custom workflow.

Note

Standard charges might apply for services deployed to your account.

Step 2: Run the state machine

After all the resources are provisioned and deployed, you can run the state machine.

1. On the State machines page, choose your sample project.

2. On the sample project page, choose Start execution.

3. In the Start execution dialog box, do the following:

   1. (Optional) Enter input values in JSON format to run your sample project.

      If you chose to Run a demo, you need not provide any execution input.

      Note

      If the demo project you deployed contains prepopulated execution input data, use that input to run the state machine.

   2. Choose Start execution.

   3. (Optional) The Step Functions console directs you to a page that's titled with your execution ID. This page is known as the Execution Details page. On this page, you can review the execution results as the execution progresses or after it's complete.

      After the execution is complete, choose individual states on the Graph view, and then choose the individual tabs on the Step details pane to view each state's details including input, output, and definition respectively.

      • For details about the execution information you can view on the Execution Details page, see Execution details overview.

      • For more information about viewing a Distributed Map state's execution in the console, see Viewing Map Runs.

   4. (Optional) Review the execution results exported to the Amazon S3 bucket. These results include data, such as execution input and output, ARN, and execution status. For more information, see ResultWriter (Map).