High Performance Architecture Lab at GT
We focus on research which enables high-performance and energy-efficient computing from microarchitectures to compilers.
Popular repositories Loading
-
CuPBoP-AMD
CuPBoP-AMD PublicCuPBoP-AMD is a CUDA translator that translates CUDA programs at NVVM IR level to HIP-compatible IR that can run on AMD GPUs.
-
-
Repositories
Showing 10 of 29 repositories
- TensorDynamic_AE Public Forked from yuxiaojia/TensorDynamic
TensorDynamic contains scripts and instructions to reproduce the key results of the paper. It includes fault injection sweeps across three backends (TensorDynamic, PyTorchFI, MRFI) on four DNN models (ResNet-20, MobileNetV2, ShuffleNetV2, YOLOv9), covering Table 2, Figure 4, Figure 5, and Figure 6.
gthparch/TensorDynamic_AE’s past year of commit activity - MacSim-User-Guide Public
gthparch/MacSim-User-Guide’s past year of commit activity - MQSim_macsim Public
gthparch/MQSim_macsim’s past year of commit activity - CuPBoP-AMD Public
CuPBoP-AMD is a CUDA translator that translates CUDA programs at NVVM IR level to HIP-compatible IR that can run on AMD GPUs.
gthparch/CuPBoP-AMD’s past year of commit activity - llm-on-cpu Public
gthparch/llm-on-cpu’s past year of commit activity