Deep Packet Inspection is an important technique for detection of network traffic anomolies.It can detect attempted or successful intrusion or legitimate traffic that causes congestion problems. There is a tradeoff between the quality of analysis and the power of the computing resources required. It is often carried out at major network hubs, where there is a very large flow of traffic. Thus, to reach its full usefulness, the analysis must be completed in real-time. The same applies to many other cybersecurity related applications.
The tremendous computing power of General-Purpose Graphics Processing Units (GPGPUs or simply GPUs) have the potential to overcome this problem; however, their efficient programming is a non-trivial exercise. This project will investigate accelerating Deep Packet Inspection or other cybersecurity-related applications on GPUs.
This project will investigate how selected parts of a suitable cybersecurity-related application may be implemented an a GPGPU using CUDA or OpenACC. comparing various techniques. In particular, it will involve: setting up and profiling a representative workload; analyzing and isolating kernels which would be fruitful for GPU implementation; implementing these kernels, evaluating their performance. Developing test harnesses for these kernels is highly desirable; this and developing methodologies for this kind of development could be used to strengthen the project's software engineering emphasis.
The PhD/MPhil version of this project would involve categorizing a broad range cybersecurity-related applications which are compute- and/or memory intensive. It would evaluate their performance on various current GPUs and identify, and if possible propose new, design features of GPUs that would be broadly suitable for the area of cybersecurity.
Note: prior GPU experience is recommended for 12-unit project courses
GPUs are a hot technology; cybersecurity is a hot field. RSCS has strong links with ASD, IBM and NVIDIA, all of which have a strong interest in cybersecuiry.