Přeskočit na obsah

Hadoop

Hadoop is a popular open-source framework for distributed storage and distributed processing of large volumes of data, especially where the processing involves a MapReduce algorithm. Unlike traditional cluster resources in MetaCentrum, which assume just-in-time stage-in of data before processing, the Hadoop environment is intended for mid- to long-term storage/collection of data to be processed.
Aside of a large native Hadoop cluster extensible on demand with additional virtual nodes, MetaCentrum can also offer purely virtual Hadoop PaaS clusters for prototyping your solutions prior to large-scale deployment.

How to Enroll

The service is intended for MetaCentrum users and registration is subject to approval:

  1. Register with MetaCentrum (unless you are already a memeber).
  2. Apply for access to Hadoop.
  3. Access the Hadoop frontend at hador.ics.muni.cz with ssh.

User Documentation and Support

Current Hardware Resources

  • 27 ⨉ 16 cores
  • 27 ⨉ 128 GB of RAM
  • 1 PB of storage in HDFS
Last changed:Mon Apr 04 14:15:48 CEST 2016