HADOOP BLOCK PLACEMENT POLICY FOR DIFFERENT FILE FORMATS
Suhas V. Ambade, Priya Deshpande · 2014
Now a day’s Peta-Bytes of data becomes the norm in industries. Handling, analyzing such big data is challenging task. Even frameworks like Hadoop (Open Source Implementation of MapReduce Paradigm) and NoSQL databases like Cassandra, HBase can be used to analyze and store such large data; heterogeneity of data is still an issue. Data centers usually have clusters formed using heterogeneous nodes. Ecosystem like Hadoop can be used to manage such types of cluster but it can not schedule jobs (Application) efficiently on this heterogeneous cluster when data is itself heterogeneous. Heterogeneity of data may be because of data format or because of the complexity. This paper is review of systems and algorithms for distributed management of huge size data, efficiency of these approaches.