DynamoDB Storage

DynamoDB storage provides a scalable and persistent solution for storing conversation history in the Multi-Agent Orchestrator System. This option is ideal for production environments where long-term data retention and high availability are crucial.

Features

• Persistent storage across application restarts
• Scalable to handle large volumes of conversation data
• Integrated with AWS services for robust security and management

When to Use DynamoDB Storage

• In production environments
• When long-term persistence of conversation history is required
• For applications that need to scale horizontally

Implementation

To use DynamoDB storage in your Multi-Agent Orchestrator:

1. Set up a DynamoDB table with the following schema:

   • Partition Key: PK (String)
   • Sort Key: SK (String)
   • Additionally, you can also set up your DynamoDB table with a TTL Key to automatically delete older conversation items

2. Use the DynamoDbChatStorage when creating your orchestrator:

TypeScript

import { DynamoDbChatStorage, MultiAgentOrchestrator } from 'multi-agent-orchestrator';

const tableName = 'YourDynamoDBTableName';
const region = 'your-aws-region';
const TTL_DURATION = 3600; // in seconds
const dynamoDbStorage = new DynamoDbChatStorage(tableName, region, 'your-ttl-key-name', TTL_DURATION);
const orchestrator = new MultiAgentOrchestrator({
  storage: dynamoDbStorage
});

Python

from multi_agent_orchestrator.storage import DynamoDbChatStorage
from multi_agent_orchestrator.orchestrator import MultiAgentOrchestrator

table_name = 'YourDynamoDBTableName'
region = 'your-aws-region'
TTL_DURATION = 3600  # in seconds
dynamodb_storage = DynamoDbChatStorage(table_name, region, ttl_key='your-ttl-key-name', ttl_duration=TTL_DURATION)
orchestrator = MultiAgentOrchestrator(storage=dynamodb_storage)

Configuration

Ensure your AWS credentials are properly set up and that your application has the necessary permissions to access the DynamoDB table.

Considerations

• Requires AWS account and proper IAM permissions
• May incur costs based on usage and data storage
• Read and write operations may have higher latency compared to in-memory storage

Best Practices

• Use DynamoDB storage for production deployments
• Implement proper error handling for network-related issues
• Consider implementing a caching layer for frequently accessed data to optimize performance
• Regularly backup your DynamoDB table to prevent data loss

DynamoDB storage offers a robust and scalable solution for managing conversation history in production environments. It ensures data persistence and allows your Multi-Agent Orchestrator System to handle large-scale deployments with reliable data storage and retrieval capabilities.