Cluster planning:
- Small cluster (2-10 nodes): Clusters of three or more machines typically use a dedicated NameNode/ResourceManager, and all other nodes are workers.
- Medium cluster (10-40 nodes): separate machines for master, usually dedicated edge node
- Large cluster (> 40 nodes): occupies multiple racks, requires individual planning
Storage layer for HDFS:
- Cloud and virtualization solutions are very popular in the enterprise world
- With Hadoop, focus on using JDOD (just bunch of disks) instead of SAN (storage area network)
- Hadoop heterogeneous storage was introduced in Hadoop 2.5
Commodity hardware:
Balanced configuration for one worker recommended by vendors:- 2-4 8-cores CPU
- 128 GB RAM
- 12 hard drives (1 or 2 TB each)
Немає коментарів:
Дописати коментар