Characteristics of Big Data Application Platform
Nati Shalom describes the main characterstics of a future Big Data platform (as compared to existing application platforms like JavaEE):
- Support Batch and Real Time analytics
- Domain model and Data access API
- Business logic
- Support new semantics that fit the dynamic web era
- Provide built-in semantics for handling of the tradeoffs between consistency, availability, scalability rather than trying to force a least common denominator as with XA and JTA
- Provide built-in support for event driven data distribution using pub/sub model
- Built in support for public/private cloud
- Open & consistent management and orchestration across the stack
While having around such a platform sounds compelling, we shouldn’t forget that some of the fundamental parts are still under development or are completely missing. Before having an integrated uniform Big Data application platform we should at least attempt to have the right building blocks and ensure that they are created with integration in mind. Even if their (customer acquisition budgets) allow Spring Data and Hibernate OGM to already work towards unifying data access layers, the lack of real scenarios and the lack of integration between storage and processing solutions might prove them to be too early to market. Anyways it is good to know that when things will start to settle and a much clearer perspective on a Big Data application platform will be available, they will already have enough experience to provide us with the right solutions.
Original title and link: Characteristics of Big Data Application Platform (©myNoSQL)
via: http://natishalom.typepad.com/nati_shaloms_blog/2011/09/big-data-application-platform.html
