Dr Zheng WangLecturer
Dr. Zheng Wang was appointed as a Lecturer in the School of Computing and Communications at Lancaster University in September, 2013. Before joining Lancaster, he was a Research Associate at the University of Edinburgh, following the completion of his PhD in 2011. Prior to that he was a research staff member in IBM Research.
Dr. Wang's research focus is in the areas of parallel compilers, runtime systems and the application of machine learning to tackle the challenging optimisation problems within these areas. My research interests include:
- Compiler-based parallelism mapping: how to map a parallel program to the underlying hardware to achieve the best (energy-efficient) performance.
- Code generation and optimisation for heterogeneous many-cores: I am interested in how can compilers generate efficient code for the emerging heterogeneous many-core systems, such as a CPU-GPGPU system.
- Auto-parallelising compilers: I am currently investigating the use of dynamic analysis together with machine learning to develop a new approach that gives scalable performance for many-cores.
- Runtime scheduling: how to schedule multiple concurrently running tasks in a multi-tasking environment to maximise the system-level performance (e.g. throughput or energy-efficiency).
- Research into energy efficient computing through system software including just-in-time compilers and operating systems
For more information, please refer to my homepage at:http://www.lancaster.ac.uk/staff/wangz3/
PhD in Computer Science, University of Edinburgh, UK
PGCAP Certificate, Lancaster University
Lecturer, School of Computing and Communications, Lancaster University, Sept. 2013 - Present
Research Fellow, School of Informatics, University of Edinburgh, 2011-2013
Research Staff member, IBM China Research Lab, 2005-2007
PhD Supervision Interests
I'm interested in supervising students in any of the areas listed below:Many-core SystemsCompiler OptimisationPower and Energy OptimisationRuntime Adaptation and Dynamic Runtime OptimisationHeterogeneous Parallelism OptimisationGPGPU OptimisationAuto-tuning and Machine Learning Techniques
Selected Publications Show all 55 publications
Fast automatic heuristic construction using active learning
Ogilvie, W., Petoumenos, P., Wang, Z., Leather, H. 09/2014 In: Languages and Compilers for Parallel Computing. Springer p. 146-160. 15 p.
Exploitation of GPUs for the parallelisation of probably parallel legacy code
Wang, Z., Powell, D., Franke, B., O'Boyle, M. 2014 In: Compiler construction. Berlin : Springer Verlag p. 154-173. 20 p.
Integrating profile-driven parallelism detection and machine-learning based mapping
Wang, Z., Tournavitis, G., Franke, B., O'Boyle, M. 2014 In: ACM Transactions on Architecture and Code Optimization. 11, 1
Using machine learning to partition streaming programs
Wang, Z., O'Boyle, M. 09/2013 In: ACM Transactions on Architecture and Code Optimization. 10, 3, 25 p.
Partitioning streaming parallelism for multi-cores: a machine learning based approach
Wang, Z., O'Boyle, M.F. 2010 In: Proceedings of the 19th International Conference on Parallel Architectures and Compilation Techniques (PACT 2010). New York, NY, USA : ACM p. 307-318. 12 p.
OpenCL task partitioning in the presence of GPU contention
Grewe, D., Wang, Z., O'Boyle, M. 2014 In: Languages and compilers for parallel computing. Springer p. 87-101. 15 p.
Portable mapping of data parallel programs to OpenCL for heterogeneous systems
Grewe, D., Wang, Z., O'Boyle, M. 2013 In: Code Generation and Optimization (CGO), 2013 IEEE/ACM International Symposium on. IEEE p. 1-10. 10 p.
Smart, adaptive mapping of parallelism in the presence of external workload
Emani, M.K., Wang, Z., O'Boyle, M. 02/2013 In: 2013 International Symposium on Code Generation and Optimization (CGO). IEEE p. 1-10. 10 p.
A workload-aware mapping approach for data-parallel programs
Grewe, D., Wang, Z., O'Boyle, M. 2011 In: HiPEAC '11 Proceedings of the 6th International Conference on High Performance and Embedded Architectures and Compilers. New York : ACM p. 117-126. 10 p.
Towards a holistic approach to auto-parallelization: integrating profile-driven parallelism detection and machine-learning based mapping
Tournavitis, G., Wang, Z., Franke, B., O'Boyle, M.F. 2009 In: Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2009). New York, NY, USA : ACM p. 177-187. 11 p.
Mapping parallelism to multi-cores: a machine learning based approach
Wang, Z., O'Boyle, M.F. 2009 In: Proceedings of the 14th ACM SIGPLAN symposium on Principles and Practice of Parallel Programming (PPoPP '09). New York, NY, USA : ACM p. 75-84. 10 p.
EPSRC: Abstraction-Level Energy Accounting and Optimisation in Many-core Programming Languages (ALEA)
31/12/2013 → 30/12/2016
- SCC Distributed Systems Group (MetaLab)