Move to managed analytics
summary
Shows how to replace self‑managed Kafka, Elasticsearch and Hadoop stacks with concrete AWS services and outcomes, not just promises. Migrating Kafka uses Amazon MSK and MirrorMaker to replicate clusters in days or weeks. Elasticsearch gains UltraWarm to turn 14 days of hot logs into 66 days for the same cost. EMR runs Spark workloads up to 2.4× faster, HBase persists on S3, and the demo uses EMR Studio to parameterize a notebook, execute it on a cluster and write the weather map output to S3.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 1:11The explosion of data
- 2:35New data sources
- 3:42AWS Lake House
- 5:50AMER Updates
- 7:42Selfmanaging analytics
- 9:33Move to managed analytics
- 10:36Amazon Kafka
- 11:38Mirrormaker
- 12:34Benefits
- 14:59Realtime analytics
- 15:35Elastic Search
- 16:38Ultra Warm
- 22:38Git Repository
- 29:48Closing notes
related talks
32mMigrating to managed Analytics services
35mMigrating to Managed Analytics Services
26mRe-architecting the Data Platform with Spark on Kubernetes
22mFireside Chat with AWS
22mHow Rocket Companies modernized their data strategy on AWS
28mInnovations in AWS analytics: Data processing
39mMigrating a data stack from AWS to Azure #bbuzz
14mLogging Apache Spark - How we made it easy