Jump to: navigation, search


< Sahara
Revision as of 09:54, 17 June 2013 by Sergey Lukjanov (talk | contribs) (Phase 1 - Basic Cluster Provisioning (completion - early April))

Phase 1 - Basic Cluster Provisioning (April, 10 - released)

  • Cluster provisioning
  • Deployment Engine implementation for pre-installed images
  • Templates for Hadoop cluster configuration
  • REST API for cluster startup and operations
  • UI integrated into Horizon

Phase 2 - Cluster Operations (completion - end of June)

  • Manual cluster scaling (add/remove nodes)
  • Hadoop cluster topology configuration parameters
    • Data node placement control
    • HDFS location
    • Swift integration
  • Integration with vendor specific deployment/management tooling
  • Monitoring support - integration with 3rd-party monitoring tools (Zabbix, Nagios)

Phase 3 - Analytics as a Service (completion - end of September)

  • API to execute Map/Reduce jobs without exposing details of underlying infrastructure (similar to AWS EMR)
  • User-friendly UI for ad-hoc analytics queries based on Hive or Pig

Further Roadmap (completion - TBD)

  • HDFS and Swift integration
    • Caching of Swift data on HDFS
    • Avoid issues with Swift eventual consistency while running job
  • HBase support