Tuesday, March 10, 2015

Service Principals


Standardized Service Contract
Grammar (XSD, XML schema), WSDL, WS standard and policy

Service Loose Coupling

Independent design of service, logic of service, dependency between two services should not exist.

Service Abstraction

Hide as much as underlying detail of the service implementation.

Service Re-usability

Reuse of the service is key objective while developing service. This is more enterprise wide.

Service Autonomy

This gives more control to the service.  It is also compliant to loose coupling principal. It is more independent of service building.

Service Statelessness

Service should be more stateless and it can be stateful only if needed. Stateless will make sure that services should release resources quickly.

Service Discoverability

Easy to understand (Interpretability), easy to discover (search).

Service Composability

Complex service needs change with time, dividing into the small problems and compose into the service so maintainability will easy.

Service-Orientation and Inter-operable

Services should be Inter-operable.  This is apply for all above. Every principal should follow this one too.



Friday, October 31, 2014

HADOOP, HFDS, MAP reduces


Hadoop and uses


There are lot of products that are based on Hadoop and used for different purposes such as SQL query, hadoop language, some of them are:

HIVE: A plug-in for Hadoop using which one can write SQL like queries on Hadoop. Hive internally converts them to MapReduce jobs.

PIG: Pig is Hadoop based language that is capable of handling very deep and very long pipelines, which is not possible in SQL.

HBASE: HBase is a non-relational database built on top of Hadoop framework. It allows quick/non-sequential data access. It also has transnational capabilities that are missing in Hadoop.

SQOOP: It is a tool that is used import data into Hadoop cluster from external sources such as Relational Database, Data warehouse etc.

Tuesday, April 1, 2014