Understanding throughput-oriented architectures
- 1 November 2010
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in Communications of the ACM
- Vol. 53 (11), 58-66
- https://doi.org/10.1145/1839676.1839694
Abstract
For workloads with abundant parallelism, GPUs deliver higher peak computational throughput than latency-oriented CPUs.Keywords
This publication has 22 references indexed in Scilit:
- Implementing sparse matrix-vector multiplication on throughput-oriented processorsPublished by Association for Computing Machinery (ACM) ,2009
- Scalable Parallel Programming with CUDAQueue, 2008
- Niagara: A 32-Way Multithreaded Sparc ProcessorIEEE Micro, 2005
- The Vector-Thread ArchitectureACM SIGARCH Computer Architecture News, 2004
- A survey of processors with explicit multithreadingACM Computing Surveys, 2003
- Vector architecturesPublished by Association for Computing Machinery (ACM) ,1998
- Exploiting heterogeneous parallelism on a multithreaded multiprocessorPublished by Association for Computing Machinery (ACM) ,1992
- The Tera computer systemPublished by Association for Computing Machinery (ACM) ,1990
- The CRAY-1 computer systemCommunications of the ACM, 1978
- Merging with parallel processorsCommunications of the ACM, 1975