May 08, 2020

AWS Lambda Functions for Python and Ruby

I love to program in Ruby as well as Python, and AWS Lambda functions provide the perfect solution to combine both languages' capabilities without additional server configuration. The aws-sdk-lambda library makes serverless computing workflow dead simple, by providing a gateway that connects to a Lambda function that will run Python code and return the results to a Rails application.

"Run code without thinking about servers. Pay only for the compute time you consume." – AWS

The primary motive for integrating Python is that it provides a rich set of Machine Learning tools to analyze Real Estate data. Much of what I wanted to accomplish, I could have with one script in a Jupyter Notebook, but integrating that functionality at scale would require a lot of overhead and building another API.

AWS does much of the heavy-lifting tasks like server provisioning and management, which can be monitored through their web interface.

Lambda passes an event parameter (usually of the Python dict type) as well as a context object to the handler—providing methods and properties with information about the invocation, function, and execution environment.

import json
import scipy

def lambda_handler(event, context):
    return {
        'statusCode': 200,
        'body': {}
    }

Here are the simple steps to get up and running with AWS:

1. Install the aws-cli

$ brew install aws-cli
$ aws configure
AWS Access Key ID: <access_key_id>
AWS Secret Access Key: <secret_access_key>
Default region name [us-west-2]: <default_region>
Default output format [None]: json
$ aws lambda list-functions

2. Add aws-sdk-lambda to Gemfile

gem 'aws-sdk-lambda', '~> 1.4', require: false

3. Connect to Aws::Lambda::Client

module Lambda
  class Python
    require 'aws-sdk-lambda'
    include Service

    REGION = 'us-east-1'.freeze

    def initialize(**params)
      @params = params
    end

    def call
      response = client.invoke(options)
      response_payload = JSON.parse(
        response.payload.string,
        symbolize_names: true
      )
      JSON.parse(response_payload[:body])
    end

    private

    def request_object
      @request_object ||= Listing.lambda_object.to_json
    end

    def permitted_params
      {
        SortBy: 'time',
        SortOrder: 'descending',
        NumberToGet: 10,
        CustomObject: request_object
      }
    end

    def payload
      @payload ||= JSON.generate(permitted_params)
    end

    def options
      {
        function_name: 'PythonClient',
        invocation_type: 'RequestResponse',
        log_type: 'None',
        payload: payload
      }
    end

    def client
      @client ||= Aws::Lambda::Client.new(
        region: REGION,
        access_key_id: :access_key_id,
        secret_access_key: :secret_access_key
      )
    end
  end
end

This configuration opens up many doors to fully leverage what Ruby and Python have to offer. As an added benefit, you are only charged for every 100ms that your code executes, saving you unnecessary compute time costs. Provisioned Concurrency keeps your functions initialized and responsive within double-digit milliseconds.

AWS Lambda enables the use of Python's powerful libraries to validate, normalize, and analyze property data before it is moved to a data store, all while maintaining programmer happiness with Ruby on Rails.