Project reference: 1615
An important part of HPC is understanding how optimisation carried out on application codes influences performance. Recently focus has shifted towards energy efficiency in the onset of “green” HPC clusters. At ICHEC we our continuously pushing the boundaries towards this goal. Our recent endeavour involves developing a special prototype system that can measure power consumption of different many-core technologies (e.g. GPUs, Intel Xeon Phi co-processors, FPGAs) used by many HPC applications. To make this prototype accessible to a wider audience this project will aim to build a performance analytics dashboard that potential developers can use to submit and/or analyse their applications using a variety of metrics such as power consumption and performance. The dashboard could also be extended to suit a larger compute cluster and to monitor further information about application behaviours.
The project will focus on developing a front end visualisation tool accessible over the web that can present the set of metrics in such a way that makes it easy for the user to analyse the relevant performance data interactively. Also some work in the backend on how different applications can be executed and the collection and storage of the performance data would need to be developed. The project would suit candidates who share an interest in HPC and visualisation alike.
Project Mentor: Dr. Michael Lysaght
Site Co-ordinator: Simon Wong
Student: Jiri Blahos
Students will learn about HPC performance considerations including energy efficiency concepts that are key to the future of HPC. The student will also learn about technologies and tools used to build the platform, e.g. web technologies, and the R statistical package and relevant libraries.
Student Prerequisites (compulsory):
Basic Linux skills; basic scripting (e.g. shell, Python, Perl); knowledge of basic programming concepts; fundamental knowledge of web development concepts (e.g. HTML, CSS) and a flair for visualisation.
Student Prerequisites (desirable):
Knowledge about application performance; experience with Python and/or the R statistical software package; prior experience with building web sites.
Week 1: Training
Week 2-3: Induction and developing project plan
Week 4-5: Developing first prototype (simple analytics interface)
Week 6-7: Refinement and further development (extend to larger clusters)
Week 8: Project write-up and demonstration
Final Product Description:
Interactive performance and energy efficiency analysis on a web page, where the user can modify the information to be displayed in real-time, including automatic generation of graphs, charts, etc.
Adapting the Project: Increasing the Difficulty:
Interested students could learn about carrying out some of the performance analysis benchmarks.
Students can carry out this work on a commodity laptop (can bring his/her own). It will also require using open source software such as Python/R.