Jeff Duffy

LLM inference depends on limited GPU memory, and a KV cache lets agentic AI applications reuse computed intermediate states across prompts to reduce latency and costs.

Join this session to learn how the new multi-Region strong consistency mode for Amazon DynamoDB global tables helps you build highly resilient applications across Regions.

NoSQL databases are purpose-built for specific data models and optimized for modern applications like mobile, web, and gaming applications that require scalability, low latency, and flexibility. AWS offers several fully managed and serverless NoSQL database services, including Amazon DynamoDB, Amazon DocumentDB (with MongoDB compatibility), and Amazon Keyspaces (for Apache Cassandra).

Using the databases that are best-suited for your workload is a key tenet of the AWS purpose-built database approach. In this session, you learn how to use Amazon DocumentDB (with MongoDB compatibility) change streams to stream data to Amazon S3 for analytics, to Amazon Elasticsearch Service for enhanced search capabilities, and to Amazon ElastiCache to pre-warm a Redis cache.

Amazon DocumentDB (with MongoDB compatibility) is a fast, reliable, fully managed MongoDB-compatible database service. What are the best practices to move your workloads to Amazon DocumentDB?

Amazon DocumentDB (with MongoDB compatibility) is a fast, scalable, highly available, and fully managed document database service that supports MongoDB workloads. In this tech talk, we will introduce Amazon DocumentDB and its unique architecture that makes it easy to run, manage & scale...