accelerators 开源项目推荐（88 个）

ucb-bar /chipyard

具有有序核心、无序核心、加速器等的敏捷RISC-V SOC设计框架

sld-columbia /esp

Embedded Scalable Platforms: Heterogeneous SoC architecture and IP integration made easy

cornell-zhang /heterocl

HeteroCL: A Multi-Paradigm Programming Infrastructure for Software-Defined Heterogeneous Computing (FPGA'19 Best Paper)

beehive-lab /jitllm

Java Inference Tornado toolkit: High performance LLM inference & serving for the JVM on GPUs

abs-tudelft /fletcher

Fletcher: A framework to integrate FPGA accelerators with Apache Arrow

areuther /ai-accelerators

CSV spreadsheets and other material for AI accelerator survey papers

Xtra-Computing /ThunderGP

HLS-based Graph Processing Framework on FPGAs

triSYCL /sycl

SYCL for Vitis: Experimental fusion of triSYCL with Intel SYCL oneAPI DPC++ up-streaming effort into Clang/LLVM

intel /qpl

Intel® Query Processing Library (Intel® QPL)

intel /DML

Intel® Data Mover Library (Intel® DML)

SET-Scheduling-Project /SET-ISCA2023

The framework for the paper "Inter-layer Scheduling Space Definition and Exploration for Tiled Accelerators" in ISCA 2023.

NVIDIA /cluster-readiness-engine

NVIDIA Cluster Readiness Engine

jha-lab /acceltran

[TCAD'23] AccelTran: A Sparsity-Aware Accelerator for Transformers

meton-robean /Vector_MulAdd_Accelerator

vector multiplication adder accelerator (using chisel 3 and RocketChip RoCC ) 向量乘法累加加速器

MPSLab-ASU /dMazeRunner

dMazeRunner: Dataflow acceleration optimization infrastructure for coarse-grained programmable accelerators

utcs-scea /ava

Automatic virtualization of (general) accelerators.

moezdil /siltide

Terminal fleet monitor for GPUs, NPUs, and 13 other accelerator kinds, from utilization down to which pod is using them.

chenxuhao /gardenia

GARDENIA: Graph Analytics Repository for Designing Efficient Next-generation Accelerators

thesis-nozal /PhD

"Optimizing Performance and Energy Efficiency in Massively Parallel Systems" PhD Dissertation repository.

AccelProf /AccelProf

A modular program analysis tool framework for accelerators (NVIDIA, AMD, and DL workloads).