Title | ||
---|---|---|
Processing-in-Memory for Energy-Efficient Neural Network Training - A Heterogeneous Approach. |
Abstract | ||
---|---|---|
Neural networks (NNs) have been adopted in a wide range of application domains, such as image classification, speech recognition, object detection, and computer vision. However, training NNs - especially deep neural networks (DNNs) - can be energy and time consuming, because of frequent data movement between processor and memory. Furthermore, training involves massive fine-grained operations with various computation and memory access characteristics. Exploiting high parallelism with such diverse operations is challenging. To address these challenges, we propose a software/hardware co-design of heterogeneous processing-in-memory (PIM) system. Our hardware design incorporates hundreds of fix-function arithmetic units and ARM-based programmable cores on the logic layer of a 3D die-stacked memory to form a heterogeneous PIM architecture attached to CPU. Our software design offers a programming model and a runtime system that program, offload, and schedule various NN training operations across compute resources provided by CPU and heterogeneous PIM. By extending the OpenCL programming model and employing a hardware heterogeneity-aware runtime system, we enable high program portability and easy program maintenance across various heterogeneous hardware, optimize system energy efficiency, and improve hardware utilization.
|
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/MICRO.2018.00059 | MICRO |
Keywords | Field | DocType |
Processing in memory,Deep neural networks,Energy efficient,Heterogeneous architecture,Programming model | Object detection,Computer architecture,Software design,Programming paradigm,Computer science,Efficient energy use,Parallel computing,Software,Software portability,Artificial neural network,Runtime system | Conference |
ISBN | Citations | PageRank |
978-1-5386-6241-0 | 12 | 0.46 |
References | Authors | |
9 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jiawen Liu | 1 | 17 | 3.24 |
Hengyu Zhao | 2 | 15 | 3.20 |
Matheus A. Ogleari | 3 | 12 | 0.46 |
Li, Dong | 4 | 764 | 48.56 |
Jishen Zhao | 5 | 638 | 38.51 |