EE596 Lab 3 -- DynamoDB

Course Webpage: EE596 -- Conversational Artificial Intelligence

Taks 1: Create and Query a DyanmoDB table

• Please follow the Getting Start for a quick walkthrough on DynamoDB.

Taks 2: Save conversation logs and user information in DynamoDB

• Create a DynamoDB table for storing conversation logs
  • Table Name: EE596Lab3ChatData
  • Partition Key: sessionId (String)
  • Sort Key: turnIdx (Number)
• Create another DynamoDB table for storing user information
  • Table Name: EE596Lab3UserData
  • Partition Key: userId (String)
• Clone the repository

  $ git clone https://github.com/hao-fang/ee596_spr2018_lab3.git
  

• Create a new Alexa Lambda function ee596_spr2018_lab3 and upload the zip file to this Lambda function. (see Lab 1 -- Task 2 for steps).

  $ cd lambdaFunc
  $ zip -r ../lambdaFunc.zip *
  

  You may also use the script upload_lambda_func.sh.
• You need to grant DynamoDB access to your Lambda function.
  • In your AWS Management Console, go to IAM dashboard.
  • In the section Roles, you can find the role you created for your lambda function (lambda_basic_execution). If you don't know which role you used for your lambda function, go to your lambda function page for ee596_spr2018_lab3 and go the section Execution role.
  • Click the lambda_basic_execution role, and choose Attach policy.
  • Search for AmazonDynamoDBFullAccess, and attach it to the role.
  • This allows your lambda function to access your DynamoDB tables.
• Change your Lab 2 Alexa Skill's Endpoint ARN to this new Lambda Function.
• Now talk to your Alexa Skill and monitor the two DynamoDB tables.

Task 3: (Optional) Design and create DynamoDB tables for your project

In this task, you will create DynamoDB tables for your project based on what you learned from Task 1 and Task 2. Explain how you plan to save data and query the tables.

Lab Checkoff

• Task 1:
  • Show the DynamoDB table you created.
  • Search for a data entry using the query operation.
• Task 2:
  • Show the DynamoDB tables you created.
  • Explain the saveSessionTurn, saveUserData, getUserData functions in lambdaFunc/index.js.
  • Make sure you know how to use corresponding APIs in Python.
• Task 3 (Optional):
  • Explain your plan about using DynamoDB tables for your project.
  • Create these DynamoDB tables.

Lab Report

• Explain the saveSessionTurn, saveUserData, getUserData functions in lambdaFunc/index.js.
• Explain the DynamoDB tables you plan to use for your project. Describe the partion key, the sort key (if any), and individual attributes.

Further Reading

• [Python and DynamoDB]
• [Improving Data Access with Secondary Indexes]