Learning curve with HBase deployment
summary
Flipkart describes an initial HBase deployment on CDH 4.2 where default settings cause frequent regionserver deaths from major GC pauses and master killoffs. Clusters start with about 1,200 regions per regionserver so region transitions are slow; region size is increased from 256 MB to much larger values, triggering compaction storms, and the team converges on optimal region size 10–15 GB and disables compaction.
AI-written summary · about
related talks
49mHow Flipkart runs HBase across zones and clouds seamlessly
55mApache HBase: crazy dances on the elephant back
21mCase Study: FOSS (Apache Hbase) at Flipkart
29mMulti Tenancy — Apache Hbase
51mBIG DATA STORAGE & PROCESSING: Introduction to Hadoop and Its Ecosystem
01h 02mScaling One of the Largest On-prem Managed MySQL Fleets in India
50mRunning Databases Efficiently at Scale - Case Study of Flipka...
22mAdvanced HBase Schema Design