Download or read book Client-server monitoring architecture for distributed supercomputing analytics written by Milos Bozovic and published by . This book was released on 2018 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: Machine learning is revolutionizing many industries, so health-care as well. Since it requires massive computing, one of the possible solutions to achieve such computing capacity is via distributed supercomputing. This concept of distributed supercomputing is the core and foundation of a research project established at Medical University in Vienna with the goal to develop a software responsible for the evaluation of high-precision diagnostic predictions. This thesis is conducted as part of the project with the goal to create a real-time analytic tool. The distributed supercomputing is conducted for multiple clients such as high-performance CPUs. Monitoring client workloads, such as CPU and memory usage, and enabling server-side data analysis is the main goal of this thesis. The developed tool shall run on a range of a client application and shall upload data to a server. The server logs a certain length of events and analyses the collected data, which can be further visualized on a corresponding web application. Web application development is beyond the scope of the thesis and will be implemented in collaboration with the author and supervisors of this thesis.*****Machine learning is revolutionizing many industries, so health-care as well. Since it requires massive computing, one of the possible solutions to achieve such computing capacity is via distributed supercomputing. This concept of distributed supercomputing is the core and foundation of a research project established at Medical University in Vienna with the goal to develop a software responsible for the evaluation of high-precision diagnostic predictions. This thesis is conducted as part of the project with the goal to create a real-time analytic tool. The distributed supercomputing is conducted for multiple clients such as high-performance CPUs. Monitoring client workloads, such as CPU and memory usage, and enabling server-side data analysis is the main goal of this thesis. The developed tool shall run on a range of a client application and shall upload data to a server. The server logs a certain length of events and analyses the collected data, which can be further visualized on a corresponding web application. Web application development is beyond the scope of the thesis and will be implemented in collaboration with the author and supervisors of this thesis.