Big Data Systems Performance: The Little Shop of Horrors
summary
Measured benchmarks compare MapReduce, in-memory engines, Flink, SAP HANA, and Oracle to show where each performs best. Results highlight huge slowdowns for disk-bound joins and demonstrate that combining compression with in-memory storage can speed analytics by an order of magnitude; examples include push-down filters and sensor telemetry clustering to illustrate when stream processing, columnar stores, or traditional warehouses succeed or fail.
AI-written summary · about
chapters (6)
- 0:00Big Data performance intro
- 7:39Data management core concepts
- 15:12Real-world data processing
- 24:19Big Data software platforms
- 33:23Performance optimization trends
- 37:13Q&A
related talks
15mOptimizing Query Performance in Big Data Systems
47mWhat's Ahead for Big Data?
45mFrom Big Data to Big Information
50mFast Analytics On Big Data
58mUnderstanding Big Data for Software Engineers
42mContemporary Approaches to Data at Scale
45mSolving Real & Big (Data) Problems Using Hadoop
36mBig Data, Disruption And The 800 Pound Gorilla In The Corner