HBase Hong Kong and China

HBase Hong Kong and China

By Sonja & Roland

[HBase Hong Kong & China

HBase is designed to host very large tables, billions of rows X millions of columns, atop clusters of industrial standard servers.  It is a distributed, versioned, column-oriented store. HBase provides Bigtable-like capabilities on top of Hadoop and HDFS. important features include Strongly consistent reads/writes, Automatic sharding, Automatic RegionServer failover, massively parallelized processing via MapReduce.

Cases of the HBase Application:

    •   uses Hadoop and HBase in several areas from social services to structured data storage and processing for internal use. 
    •   has about 30 nodes running Hadoop and HBase,  planning a deployment on an 80 nodes cluster. 
    •   write data constantly to HBase with MapReduce jobs to process then store it back to HBase or external systems. 
    •   the production cluster has been running since Oct 2008. 

    •  532 nodes cluster (8 * 532 cores, 5.3PB).
    •  heavy usage of Java MapReduce, Pig, Hive, HBase
    •  using it for Search optimization and Research.

eCircle (TeraData)
    •  two 60 nodes cluster, total 5T Ram, 1PB 
    •  mostly HBase, some MapReduce
    •  marketing data handling 

More cases ......


Please feel free to use the contact-us form to contact us now if you have any queries.

PostgreSQL, Open Source, database, Oracle, SQLServer, MYSQL