FPGA Accelerators for Densely Networked Systems (SAF16)

The Internet of Things is a paradigm of massive connectedness, bringing together a plethora of smart distributed sensors, and complex analytics in datacentres to solve problems from intelligent transportation to personalised medicine. However, the compute infrastructure we have at hand today is built upon decades-old assumptions about how to combine computation and communication, and suffers from fundamental inefficiencies as a result. We are interested in re-evaluating how we compute in this connected context, rethinking models from the ground up to take advantage of alternative computing architectures.

This PhD project is concerned with proposing methods and infrastructure for incorporating efficient accelerator technology on FPGAs within connected distributed computing systems. It includes work on virtualising data interfaces in hardware, integrating these accelerators within software infrastructure, and building applications to demonstrate the advantages of this approach. The focus will be on big data analytics, and the project can skew towards centralised datacentre or distributed embedded applications.

Through this PhD, you will gain experience in a currently niche area that is experiencing significant expansion and high demand with companies like Intel, Microsoft, Amazon, and Huawei, who are all exploring integration of FPGAs within their systems. You will also have the opportunity to work with international collaborators.

Applications will be considered as they arrive, in the run up to the final deadline, so please apply early. The start date can be brought forward. Informal enquiries can be directed at Dr Suhaib Fahmy ([Email Address Removed]).

Qualifications and Eligibility:

A first class Masters degree in electrical engineering or computer science, with some evidence of familiarity with digital design and low-level programming is required. Industry experience is a significant advantage as are other demonstrations of technical familiarity with FPGAs. Due to funding restrictions this award is available for well-qualified UK or EU students only.