Local and Global Measures for Measuring the Performance of Big Data Analytics (BDA) Process

Yusmadi Yah Jusoh and Ismail M


Performance measurement is defined as the process of quantifying the efficiency and effectiveness of an action or a process. BDA system, viewed a process, has both front-end and back-end performance properties[1]. The BDA process phases, data acquisition, data preparation, and data analysis, determine how the systems performs according to a given time and resources. In visualization and interpretation phases, users see the analytics results and interpret them into their context[2].  The performance measurement of the BDA process uses two categories of measures: Local measures and global measures (refer to Figure 1). The local measures consist of efficiency measures to assess the BDA system’s functions, and effectiveness measures to assess user satisfaction towards the system. The efficiency measures include time, capacity, response time, throughput, processing time, accuracy, resource utilization, flexibility, and provenance, those scrutinize the speed and the capability of BDA system to meet user and business needs.  The effectiveness measures, on the other hand, includes measures like user satisfaction, information representation, timeliness, reliability, understandability, and customizability.

Relevant Publications in Global Educational Journal of Library and Information Science