| Summary: | Background. In today’s competitive world, dealing with real-time streaming data is a difficult task to be achieved by many organizations. The importance of real time streaming data is rapidly increasing in all software industries by passing time. For quick growth of the companies, the data should be analysed immediately as data will be changing in fraction of second. The huge data will be generated every day and it will lead to problems such as overload of resources, Performance delays etc.., Which in turn will impact behaviour of the system. Finding the problem area in real time is difficult task to achieve as the data changes every second. Dealing with detection of bottlenecks and making decisions to handle the problem area, based on the real time data has been slow over the past years. It is also complicated due to time and effort required for storing and analysing. Organizations are not intended to wait for decision making information up to weeks or months. Organizations need to make an timely-accurate decisions by detecting problem area, in real time to improve their business support systems behaviour and performance. One of the better solutions is through data visualization as an approach. The visualizations are developed and evaluated by using task based approach. The data is collected using interviews and paper survey, to obtain the effective and efficient visualization in detecting bottlenecks. Objectives. The main objective is to find the most effective and efficient data visualization technique for real time streaming data to detect potential bottlenecks. Methods. In this research study, an action research is opted to answer the objectives. We have used interviews and paper survey to collect data in the terms of performance time, accuracy rate and user preference. Data analysis is performed using the Statistical tests and Narrative analysis method. Results. The final results obtained are the effective and efficient visualization techniques based on less performance time, higher accuracy rate and better user preference. Conclusions. An effective and efficient visualization technique for detection of bottlenecks is obtained for real time streaming data. Different categories of tasks has been used to obtain accurate results.
|
|---|
