We present a DNN inference engine, called DUET, that explores potential concurrent execution opportunities on heterogeneous CPU-GPU architecture for DNN ...
We present a DNN inference engine, called DUET, that explores potential concurrent execution opportunities on heterogeneous CPU-GPU architecture for DNN ...
The experimental results show that DUET achieves 1.3–6.4 times speed-ups on CPU and 1.5–2.3 times speed-ups on GPU, respectively, on three DNNs – Wide-and-Deep ...
DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture. M. Zhang, Z. Hu, and M. Li. IPDPS, page 151-161 ...
DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture · Author(s): · Minjia Zhang · Zehua Hu · Mingqin Li.
DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture by Zhang, Minjia, Zehua Hu, and Mingqin Li ...
2021. DUET: A compiler-runtime subgraph scheduling approach for tensor programs on a coupled CPU-GPU architecture. M Zhang, Z Hu, M Li. 2021 IEEE International ...
DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture, published at IPDPS 2021. Menghao Li, intern ...
Sep 22, 2024 · DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture by Minjia et al., IEEE IPDPS 2021 ...
DUET: A Compiler-Runtime Subgraph Scheduling Approach for Tensor Programs on a Coupled CPU-GPU Architecture pp. 151-161. CAGC: A Content-aware Garbage ...