Cloud Compute Cost Showdown
summary
Claims here include 50–80% savings on EBS by using instance local storage (example: i3 4xlarge with 3.6 TB) and 25% from autoscaling tweaks by preferring static sizing when workloads are predictable. Specific tools and techniques named: Databricks Photon and Delta Lake, Spark UI tuning (shuffle partitions, caching), S3 lifecycle and vacuuming for Hudi/Iceberg, spot vs on‑demand, and EMR flavors (EC2, EKS, Serverless) with 20% from Graviton instances and automated cluster shutdowns and alerts.
AI-written summary · about
chapters (6)
- 0:00Introduction and Agenda
- 0:48Understanding Databricks
- 4:06Databricks Cost Optimization Strategies
- 10:29Introduction to AWS EMR
- 13:48AWS EMR Cost Optimization Strategies
- 17:09Conclusion and Q&A
related talks
50mThe Great Open Cloud Shootout
26mHow Cassandra 4.X Help in Cutting Cost Mission
28mInnovations in AWS analytics: Data processing
52mMore for less: Cost optimizing your Azure SQL databases
29mAccelerate your cloud architectures with the right storage options
08mCloud-Based Data Integration is Transforming Enterprise Data
40mCloud Databases: Pros and Cons
59mBuilding the Best Cloud Data Architecture: Avoiding Costly Pitfalls