FREE PhD Study Fairs in Sheffield & Edinburgh | REGISTER NOW FREE PhD Study Fairs in Sheffield & Edinburgh | REGISTER NOW

Microbenchmarks for Accelerator Offloading


   School of Informatics

   Monday, October 31, 2022  Competition Funded PhD Project (European/UK Students Only)

About the Project

The EPCC OpenMP microbenchmarks have become a de facto standard tool for measuring the performance of OpenMP runtime implementations. With the advent of higher level, portable APIs for GPU programming such as OpenMP 4.0 and later, OpenACC and SYCL, no such equivalent set of benchmarks has become widely accepted. It would also be of interest to produce microbenchmarks that permit sensible direct comparisons between different APIs as well as between implementations. The EPCC OpenMP microbenchmark methodology cannot be adopted directly due to the lack of timing routines on the accelerator devices, so a different methodology needs to be developed. 

The statistical aspects of microbenchmarking are generally poorly handled and not well understood. A more robust statistical approach should be taken, for example so that reliable error estimates of the benchmark measurements can be reported.  

Overview of research area:

Microbenchmarks are an important tool for understanding and characterising hardware and API implementation performance. They complement application-level benchmarks by isolating very specific aspects of system performance, and provide standardised metrics by which new developments (both academic and commercial) can be assessed. 

Potential research question(s)

  • How to design microbenchmarks for GPUs - what is the best methodology for separating the overheads of e.g. kernel launch from the actual computation performed when only the sum of both can be measured 
  • How to design benchmarks that make fair comparisons across different APIs 
  • What is the state of the art performance-wise in current implementations of these APIs 
  • How (and why) do multiple measurements of the same benchmark/overhead vary, and what are the best statistical practices to make accurate unbiased measurements and robust error estimates. 

Student Requirements:

Note that these are the minimum requirements to be considered for admission.

A UK 2:1 honours degree, or its international equivalent, in a relevant subject such as computer science and informatics, physics, mathematics, engineering, biology, chemistry and geosciences.

You must be a competent programmer in at least one of C or C++.  

English Language requirements as set by University of Edinburgh

Student Recommended/Desirable Skills and Experience

Experience with:

  • Computer architecture
  • OpenMP 
  • GPU programming
  • Statistical analysis and error estimation

Funding Notes

EPCC holds the following funding opportunities across its PhD opportunities at present for which this project is one of many eligible (i.e. competitive funding):
For entry during academic year 2022-23:
At least one EPSRC studentship with standard EPSRC eligibility: View Website
We also welcome applications for these projects from students with their own source(s) of funding

References

https://www.epcc.ed.ac.uk/research/computing/performance-characterisation-and-benchmarking/epcc-openmp-micro-benchmark-suite

How good is research at University of Edinburgh in Computer Science and Informatics?


Research output data provided by the Research Excellence Framework (REF)

Click here to see the results for all UK universities

Email Now


Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs