Improvisation of Incremental Computing In Hadoop Architecture- A Literature?

Automatic increment in data is a difficult problem, as it requires the development of well-defined algorithms and a runtime system to support performance code. Many online data sets grow incrementally over time as new entries are slowly added and existing entries are deleted or updated to manage the dataset. The Hadoop is a dedicated distributed paradigm used to manipulate the large amount of distributed data. This manipulation contains not only storage but also the computation and processing of the data. Hadoop is normally used for data centric applications. Systems for incremental bulk data processing and computation can efficiently use for the updates but are not compatible with the non-incremental systems such as e.g., MapReduce, and more importantly and requires the programmer to implement application-specific dynamic/ incremental algorithms, ultimately increasing algorithm and code complexity. Thus this paper discusses about the various aspects of the incremental computation.