HBASE SCHEMA DESIGN - ApacheCon

More documents

Recommendations

Info

HBase Tables and Regions • Table is made up of any number if regions • Region is specified by its startKey and endKey • Empty table: (Table, NULL, NULL) • Two-region table: (Table, NULL, “com.cloudera.www”) and (Table, “com.cloudera.www”, NULL) • Each region may live on a different node and is made up of several HDFS files and blocks, each of which is replicated by Hadoop
Distribution
Page 1 and 2: HBASE SCHEMA DESIGN and Cluster Siz
Page 3 and 4: Agenda • HBase Architecture • S
Page 5 and 6: HBase Tables
Page 7: HBase Tables
Page 11 and 12: HBase Architecture
Page 13 and 14: HBase Architecture (cont.) • Base
Page 15 and 16: Compactions • General Concepts
Page 17 and 18: Region Splits • Triggered by conf
Page 19 and 20: Auto Sharding and Distribution •
Page 21 and 22: Storage Separation • Column Famil
Page 23 and 24: SCHEMA DESIGN
Page 25 and 26: Key Cardinality • The best perfor
Page 27 and 28: Fold, Store, and Shift • Logical
Page 29 and 30: DDI • Stands for Denormalization,
Page 31 and 32: Tall-Narrow vs. Flat-Wide Tables
Page 33 and 34: Partial Key Scans Key - -- --- De
Page 35 and 36: Salting • Prefix row keys to gain
Page 37 and 38: Key Design
Page 39 and 40: Example: Facebook Insights • > 20
Page 41 and 42: Key Design • Reversed Domains •
Page 43 and 44: Summary • Design for Use-Case •
Page 45 and 46: CLUSTER SIZING
Page 47 and 48: Memory Sharing • By default every
Page 49 and 50: Block Cache • Use exported metric
Page 51 and 52: Write Performance • There are man
Page 53 and 54: Write-Ahead Log (cont.) • Size se
Page 55 and 56: Flushes • Every mutation call (pu
Page 57 and 58: Dependencies • Flushes happen acr
Page 59 and 60:
Some More Numbers • Under real wo
Page 61 and 62:
Cheat Sheet #1 • Ensure you have
Page 63 and 64:
Example • Java Xmx heap at 10GB
show all

HBASE SCHEMA DESIGN - ApacheCon

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?