Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313447739> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4313447739 abstract "The continued growth in the processing power of FPGAs coupled with high bandwidth memories (HBM), makes systems like the Xilinx U280 credible platforms for linear solvers which often dominate the run time of scientific and engineering applications. In this paper, we present Callipepla, an accelerator for a preconditioned conjugate gradient linear solver (CG). FPGA acceleration of CG faces three challenges: (1) how to support an arbitrary problem and terminate acceleration processing on the fly, (2) how to coordinate long-vector data flow among processing modules, and (3) how to save off-chip memory bandwidth and maintain double (FP64) precision accuracy. To tackle the three challenges, we present (1) a stream-centric instruction set for efficient streaming processing and control, (2) vector streaming reuse (VSR) and decentralized vector flow scheduling to coordinate vector data flow among modules and further reduce off-chip memory accesses with a double memory channel design, and (3) a mixed precision scheme to save bandwidth yet still achieve effective double precision quality solutions. To the best of our knowledge, this is the first work to introduce the concept of VSR for data reusing between on-chip modules to reduce unnecessary off-chip accesses for FPGA accelerators. We prototype the accelerator on a Xilinx U280 HBM FPGA. Our evaluation shows that compared to the Xilinx HPC product, the XcgSolver, Callipepla achieves a speedup of 3.94x, 3.36x higher throughput, and 2.94x better energy efficiency. Compared to an NVIDIA A100 GPU which has 4x the memory bandwidth of Callipepla, we still achieve 77% of its throughput with 3.34x higher energy efficiency. The code is available at https://github.com/UCLA-VAST/Callipepla." @default.
- W4313447739 created "2023-01-06" @default.
- W4313447739 creator A5016776689 @default.
- W4313447739 creator A5025422213 @default.
- W4313447739 creator A5032647120 @default.
- W4313447739 creator A5036239104 @default.
- W4313447739 creator A5060115951 @default.
- W4313447739 creator A5074835029 @default.
- W4313447739 date "2023-02-12" @default.
- W4313447739 modified "2023-09-26" @default.
- W4313447739 title "Callipepla: Stream Centric Instruction Set and Mixed Precision for Accelerating Conjugate Gradient Solver" @default.
- W4313447739 cites W1965512225 @default.
- W4313447739 cites W1969468690 @default.
- W4313447739 cites W2045285941 @default.
- W4313447739 cites W2053999255 @default.
- W4313447739 cites W2088899870 @default.
- W4313447739 cites W2094756095 @default.
- W4313447739 cites W2136897996 @default.
- W4313447739 cites W2316564661 @default.
- W4313447739 cites W2478429860 @default.
- W4313447739 cites W2585560244 @default.
- W4313447739 cites W2897997157 @default.
- W4313447739 cites W2899644485 @default.
- W4313447739 cites W2899691047 @default.
- W4313447739 cites W3008230609 @default.
- W4313447739 cites W3008730548 @default.
- W4313447739 cites W3016769527 @default.
- W4313447739 cites W3092394454 @default.
- W4313447739 cites W3106125969 @default.
- W4313447739 cites W3130920634 @default.
- W4313447739 cites W3136950882 @default.
- W4313447739 cites W3198975860 @default.
- W4313447739 cites W4281657881 @default.
- W4313447739 cites W4285049632 @default.
- W4313447739 cites W4293024216 @default.
- W4313447739 cites W4308083582 @default.
- W4313447739 doi "https://doi.org/10.1145/3543622.3573182" @default.
- W4313447739 hasPublicationYear "2023" @default.
- W4313447739 type Work @default.
- W4313447739 citedByCount "0" @default.
- W4313447739 crossrefType "proceedings-article" @default.
- W4313447739 hasAuthorship W4313447739A5016776689 @default.
- W4313447739 hasAuthorship W4313447739A5025422213 @default.
- W4313447739 hasAuthorship W4313447739A5032647120 @default.
- W4313447739 hasAuthorship W4313447739A5036239104 @default.
- W4313447739 hasAuthorship W4313447739A5060115951 @default.
- W4313447739 hasAuthorship W4313447739A5074835029 @default.
- W4313447739 hasBestOaLocation W43134477391 @default.
- W4313447739 hasConcept C107027933 @default.
- W4313447739 hasConcept C149635348 @default.
- W4313447739 hasConcept C165005293 @default.
- W4313447739 hasConcept C173608175 @default.
- W4313447739 hasConcept C188045654 @default.
- W4313447739 hasConcept C199360897 @default.
- W4313447739 hasConcept C2776257435 @default.
- W4313447739 hasConcept C2778770139 @default.
- W4313447739 hasConcept C31258907 @default.
- W4313447739 hasConcept C41008148 @default.
- W4313447739 hasConcept C42935608 @default.
- W4313447739 hasConcept C68339613 @default.
- W4313447739 hasConcept C76155785 @default.
- W4313447739 hasConcept C9390403 @default.
- W4313447739 hasConceptScore W4313447739C107027933 @default.
- W4313447739 hasConceptScore W4313447739C149635348 @default.
- W4313447739 hasConceptScore W4313447739C165005293 @default.
- W4313447739 hasConceptScore W4313447739C173608175 @default.
- W4313447739 hasConceptScore W4313447739C188045654 @default.
- W4313447739 hasConceptScore W4313447739C199360897 @default.
- W4313447739 hasConceptScore W4313447739C2776257435 @default.
- W4313447739 hasConceptScore W4313447739C2778770139 @default.
- W4313447739 hasConceptScore W4313447739C31258907 @default.
- W4313447739 hasConceptScore W4313447739C41008148 @default.
- W4313447739 hasConceptScore W4313447739C42935608 @default.
- W4313447739 hasConceptScore W4313447739C68339613 @default.
- W4313447739 hasConceptScore W4313447739C76155785 @default.
- W4313447739 hasConceptScore W4313447739C9390403 @default.
- W4313447739 hasLocation W43134477391 @default.
- W4313447739 hasLocation W43134477392 @default.
- W4313447739 hasLocation W43134477393 @default.
- W4313447739 hasOpenAccess W4313447739 @default.
- W4313447739 hasPrimaryLocation W43134477391 @default.
- W4313447739 hasRelatedWork W1509211761 @default.
- W4313447739 hasRelatedWork W1523072846 @default.
- W4313447739 hasRelatedWork W1531488649 @default.
- W4313447739 hasRelatedWork W2031026393 @default.
- W4313447739 hasRelatedWork W2133693067 @default.
- W4313447739 hasRelatedWork W2276173551 @default.
- W4313447739 hasRelatedWork W2391299576 @default.
- W4313447739 hasRelatedWork W2526107130 @default.
- W4313447739 hasRelatedWork W2791534362 @default.
- W4313447739 hasRelatedWork W2582864838 @default.
- W4313447739 isParatext "false" @default.
- W4313447739 isRetracted "false" @default.
- W4313447739 workType "article" @default.