Abstract | ||
---|---|---|
Neural networks span a wide range of applications of industrial and commercial significance. Binary neural networks (BNN) are particularly effective in trading accuracy for performance, energy efficiency, or hardware/software complexity. Here, we introduce a spintronic, re-configurable in-memory BNN accelerator, PIMBALL: Processing In Memory BNN AcceL(L)erator, which allows for massively parallel and energy efficient computation. PIMBALL is capable of being used as a standard spintronic memory (STT-MRAM) array and a computational substrate simultaneously. We evaluate PIMBALL using multiple image classifiers and a genomics kernel. Our simulation results show that PIMBALL is more energy efficient than alternative CPU-, GPU-, and FPGA-based implementations while delivering higher throughput.
|
Year | DOI | Venue |
---|---|---|
2020 | 10.1145/3357250 | ACM Transactions on Architecture and Code Optimization (TACO) |
Keywords | Field | DocType |
Processing in memory, binary neural networks, computational random access memory, non-volatile memory | Kernel (linear algebra),Efficient energy use,Computer science,Massively parallel,Parallel computing,Implementation,Non-volatile memory,Computational science,Throughput,Programming complexity,Artificial neural network | Journal |
Volume | Issue | ISSN |
16 | 4 | 1544-3566 |
Citations | PageRank | References |
4 | 0.51 | 0 |
Authors | ||
8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Salonik Resch | 1 | 8 | 4.37 |
S. Karen Khatamifard | 2 | 17 | 3.06 |
Zamshed I. Chowdhury | 3 | 15 | 4.45 |
Masoud Zabihi | 4 | 14 | 4.10 |
Zhengyang Zhao | 5 | 10 | 3.72 |
Jian-Ping Wang | 6 | 6 | 3.26 |
S. S. Sapatnekar | 7 | 396 | 27.16 |
Ulya R. Karpuzcu | 8 | 277 | 22.27 |