Skip to content

Latest commit

 

History

History
26 lines (19 loc) · 1.18 KB

cloud-bigtable.md

File metadata and controls

26 lines (19 loc) · 1.18 KB
Raw

Overview

Next

Cloud Bigtable (NoSQL; Wide-table DB)

  • Petabyte scale
  • High read write throughput at low latency
  • Supports Apache HBASE API
  • Processing is separated from storage
  • Data storage is massively scalable (scales up well)
  • Operational and Analytical apps because it supports high throughput and low latency
  • Learns to adjust to frequent access patterns and evenly distribute workloads among nodes
  • Big Table scales up well. Linear scaling throughput performance per Bigtable node added.
  • Smallest Bigtable cluster has Minimum of 3 nodes (can handle 30,000 operations/s). You pay for 3 nodes whether in use or not.

Cloud Bigtable storage model

  • Massively scalble tables
  • Key-value map

Processing is separated from storage

A Cloud Bigtable table is sharded in block of contiguous rows called Tablets to help balance the workload of queries

Tablets are stored in Colossus file system which is an SS table. An SS table provides persistent; ordered; immutable map from keys to values where both key and value are arbitrary byte strings