Customizable Heterogeneous Acceleration for Tomorrow's High-Performance Computing

High-performance computing as we know it today is experiencing unprecedented changes, encompassing all levels from technology to use cases. This paper explores the adoption of customizable, deeply heterogeneous manycore systems for future QoS-sensitive and power-efficient high-performance computing. At the heart of the proposed architecture is a NoC-based manycore system embracing medium-end CPUs, GPU-like processors, and reconfigurable hardware regions. The paper discusses the high-level design principles inspiring this innovative architecture as well as the key role that heterogeneous acceleration, ranging from multicore processors and GPUs down to FPGAs, might play for tomorrow's high-performance computing.

[1]  Alejandro Duran,et al.  The Intel® Many Integrated Core Architecture , 2012, 2012 International Conference on High Performance Computing & Simulation (HPCS).

[2]  Estela Suarez,et al.  Dynamical Exascale Entry Platform: the DEEP Project , 2011 .

[3]  Alessandro Cilardo,et al.  Interplay of loop unrolling and multidimensional memory partitioning in HLS , 2015, 2015 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[4]  Ra Inta,et al.  The "Chimera": An Off-The-Shelf CPU/GPGPU/FPGA Hybrid Computing Platform , 2012, Int. J. Reconfigurable Comput..

[5]  Alessandro Cilardo,et al.  Improving Multibank Memory Access Parallelism with Lattice-Based Partitioning , 2015, ACM Trans. Archit. Code Optim..

[6]  Denis Navarro,et al.  Janus: An FPGA-Based System for High-Performance Scientific Computing , 2007, Computing in Science & Engineering.

[7]  Cédric Bastoul,et al.  Code generation in the polyhedral model is easier than you think , 2004, Proceedings. 13th International Conference on Parallel Architecture and Compilation Techniques, 2004. PACT 2004..

[8]  Stephen Booth,et al.  Maxwell - a 64 FPGA Supercomputer , 2007, Second NASA/ESA Conference on Adaptive Hardware and Systems (AHS 2007).