Hard or/and Soft tuning of HADOOP Ecosystem towards query energy efficiency on NoSQL Database
Data Group system
With the emergence of the big data age, along with the data-centric and -intensive computing trend, the great amount of energy consumed by database systems has become a major concern in a society that pursues green information technology. Apache Hadoop framework supports the storing and processing of big data datasets using simple programming models. Today, in many compagnies Hadoop becoming the de facto standard for data collection and management. Research work on energy efficiency in the Hadoop ecosystem can be categorized into two approaches: software-based and hardware-based. The software approach covers Scheduling, data placement and resource management but does not make an in-depth analysis of the query processing system within Hadoop systems. The objective of this master thesis work is to analyze the operation of the Hadoop ecosystem during the execution of analytical queries in order to identify configurations along several dimensions (database type, storage engine, partitioning form, etc.) that can better optimize energy consumption.
Lõputöö kaitsmise aasta
Simon Pierre DEMBELE