Automatically accelerating non-numerical programs by architecture-compiler co-design
Communications of the ACM
MetadataShow full item record
Campanoni, S., Brownell, K., Kanev, S., Wei, G., Brooks, D., & Jones, T. M. (2017). Automatically accelerating non-numerical programs by architecture-compiler co-design. Communications of the ACM, 60 (12), 88-97. https://doi.org/10.1145/3139461
Because of the high cost of communication between processors, compilers that parallelize loops automatically have been forced to skip a large class of loops that are both critical to performance and rich in latent parallelism. HELIX-RC is a compiler/microprocessor co-design that opens those loops to parallelization by decoupling communication from thread execution in conventional multicore architecures. Simulations of HELIX-RC, applied to a processor with 16 Intel Atom-like cores, show an average of 6.85× performance speedup for six SPEC CINT2000 benchmarks.
External DOI: https://doi.org/10.1145/3139461
This record's URL: https://www.repository.cam.ac.uk/handle/1810/279417