Abstract | ||
---|---|---|
There is an increasing need to bring machine learning to a wide diversity of hardware devices. Current frameworks rely on vendor-specific operator libraries and optimize for a narrow range of server-class GPUs. Deploying workloads to new platforms - such as mobile phones, embedded devices, and accelerators (e.g., FPGAs, ASICs) - requires significant manual effort. We propose TVM, a compiler that exposes graph-level and operator-level optimizations to provide performance portability to deep learning workloads across diverse hardware back-ends. TVM solves optimization challenges specific to deep learning, such as high-level operator fusion, mapping to arbitrary hardware primitives, and memory latency hiding. It also automates optimization of low-level programs to hardware characteristics by employing a novel, learning-based cost modeling method for rapid exploration of code optimizations. Experimental results show that TVM delivers performance across hardware back-ends that are competitive with state-of-the-art, hand-tuned libraries for low-power CPU, mobile GPU, and server-class GPUs. We also demonstrate TVMu0027s ability to target new accelerator back-ends, such as the FPGA-based generic deep learning accelerator. The system is open sourced and in production use inside several major companies. |
Year | Venue | Field |
---|---|---|
2018 | OSDI | Computer architecture,End-to-end principle,Computer science,Field-programmable gate array,Compiler,Optimizing compiler,Operator (computer programming),Software portability,Artificial intelligence,Deep learning,CAS latency,Distributed computing |
DocType | Citations | PageRank |
Conference | 11 | 0.59 |
References | Authors | |
0 | 12 |
Name | Order | Citations | PageRank |
---|---|---|---|
Tianqi Chen | 1 | 1887 | 83.63 |
Thierry Moreau | 2 | 105 | 8.54 |
Ziheng Jiang | 3 | 67 | 7.19 |
Zheng, Lianmin | 4 | 15 | 1.71 |
Eddie Q. Yan | 5 | 44 | 3.53 |
Haichen Shen | 6 | 163 | 8.06 |
Meghan Cowan | 7 | 17 | 1.72 |
Leyuan Wang | 8 | 28 | 2.74 |
Yuwei Hu | 9 | 39 | 4.19 |
Luis Ceze | 10 | 2183 | 125.93 |
Carlos Guestrin | 11 | 9220 | 488.92 |
Arvind Krishnamurthy | 12 | 4540 | 312.24 |