ALL COVERED TOPICS

NoSQL Benchmarks NoSQL use cases NoSQL Videos NoSQL Hybrid Solutions NoSQL Presentations Big Data Hadoop MapReduce Pig Hive Flume Oozie Sqoop HDFS ZooKeeper Cascading Cascalog BigTable Cassandra HBase Hypertable Couchbase CouchDB MongoDB OrientDB RavenDB Jackrabbit Terrastore Amazon DynamoDB Redis Riak Project Voldemort Tokyo Cabinet Kyoto Cabinet memcached Amazon SimpleDB Datomic MemcacheDB M/DB GT.M Amazon Dynamo Dynomite Mnesia Yahoo! PNUTS/Sherpa Neo4j InfoGrid Sones GraphDB InfiniteGraph AllegroGraph MarkLogic Clustrix CouchDB Case Studies MongoDB Case Studies NoSQL at Adobe NoSQL at Facebook NoSQL at Twitter

NAVIGATE MAIN CATEGORIES

Close

HBase Write Ahead Log

A new HBase utility from Outerthought, HBase Rowlog library:

If you have an HBase table, and each time a create, update or delete operation happens on a row of this table an index should be updated you’ll need a system to guarantee that this update actually happens even in the case of failing nodes. A Write Ahead Log (WAL) can give you such a guarantee.

The purpose of the WAL is to guarantee the execution of “secondary, subordinate actions” in case a Lily repository node would die in the midst of their execution. These actions are synchronous actions, this means that (1) they should complete before returning execution to the caller, and (2) they should complete before the next update on the record is allowed to happen.

I think Riak solved this using pre-commit and post-commit hooks and CouchDB using the _changes feed (a basic demo of using CouchDB _changes or you can read about a generic CouchDB _changes consumer or streaming notifications about updates).

via: http://lilycms.org/lily/roadmap/sketchbook/hbaserowlog.html