Advances in GPU Research and Practice

Hamid Sarbazi-Azad is currently a professor of computer science and engineering at Sharif University of Technology, and a research professor of computer science at the Institute for Research in Fundamental Sciences, both in Tehran, Iran . His research interests include high-performance computer/memory architectures, NoCs and SoCs, learning architectures,, social networks, and storage systems, on which he has published over 400 refereed papers. He received Khwarizmi International Award in 2006, TWAS Young Scientist Award in engineering sciences in 2007, and Sharif University Distinguished Researcher awards in years 2004, 2007, 2008, 2010 and 2013. He is now an associate editor of ACM Computing Surveys, IEEE Computer Architecture Letters, and Elsevier’s Computers and Electrical Engineering.

Part 1: Programming and Tools 1. Formal analysis techniques for reliable GPU programming: current solutions and call to action 2. SnuCL: A unified OpenCL framework for heterogeneous clusters 3. Thread communication and synchronization on massively parallel GPUs 4. Software-level task scheduling on GPUs 5. Data placement on GPUs

Part 2: Algorithms and Applications 6. Biological sequence analysis on GPU 7. Graph algorithms on GPUs 8. GPU alignment of two and three sequences 9. Augmented Block Cimmino Distributed Algorithm for solving tridiagonal systems on GPU 10. GPU computing applied to linear and mixed-integer programming 11. GPU-accelerated shortest paths computations for planar graphs 12. GPU sorting algorithms 13. MPC: An effective floating-point compression algorithm for GPUs 14. Adaptive sparse matrix representation for efficient matrix-vector multiplication

Part 3: Architecture and Performance 15. A framework for accelerating bottlenecks in GPU execution with assist warps 16. Accelerating GPU accelerators through neural algorithmic transformation 17. The need for heterogeneous network-on-chip architectures with GPGPUs: A case study with photonic interconnects 18. Accurately modeling GPGPU frequency scaling with the CRISP performance model

Part 4: Power and Reliability 19. Energy and power considerations of GPUs 20. Architecting the last-level cache for GPUs using STT-MRAM nonvolatile memory 21. Power management of mobile GPUs 22. Advances in GPU reliability research 23. Addressing hardware reliability challenges in general-purpose GPUs