Performance Data Analysis for Parallel Processing Using Bigdata Distribution

Abstract:

The following document presents metrics and pointers for datacenter performance evaluation, whose production workflow will be improved by a parallel computing software, each cluster instance was virtualized providing for scalability and availability for every person who access to the system at different locations. Apache spark will be used as parallel processing distribution through different scenarios, each one will handle workload on physical and virtual nodes, after the collection of time response a comparations will be realized for determinate if the parallel distribution is an ideal solution for guarantee processing requirements.

Año de publicación:

2019

Keywords:

• Parallel computing
• Performance Analysis
• scalability
• quality of services

Fuente:

scopus

google