joshita / dev

Published

- 1 min read

HBase

HBase Nosql Big Data Scala/Java PySpark
img of HBase

Hbase Details

  1. Hbase is a key value columnar database
  2. Provides strong consistency (WAL)
  3. Built on top of hadoop, so there is latency as it works on map reduce
  4. Lacks a query language, used API or mapreduce
  5. replication is managed by hdfs , 3x
  6. Data backup: relies on hdfc snapshots
  7. optimized for sparse dataset and analytical workload, ideal for random reads like analytics or iot applications
  8. adhoc queries / random reads and writes
  9. natural choice for systems with mapreduce and hadoop integrations
  10. Master slave architecture
  11. Used with query engine like apache phoenix
  12. Low latency reads, because a key value store