Name
Abstract | ||
|---|---|---|
We tackle the important problem class of solving nonlinear partial differential equations. While nonlinear PDEs are typically solved in high-performance supercomputers, they are increasingly used in graphics and embedded systems, where efficiency is important.
We use a hybrid analog-digital computer architecture to solve nonlinear PDEs that draws on the strengths of each model of computation and avoids their weaknesses. A weakness of digital methods for solving nonlinear PDEs is they may not converge unless a good initial guess is used to seed the solution. A weakness of analog is it cannot produce high accuracy results. In our hybrid method we seed the digital solver with a high-quality guess from the analog side.
With a physically prototyped analog accelerator, we use this hybrid analog-digital method to solve the two-dimensional viscous Burgers' equation ---an important and representative PDE. For large grid sizes and nonlinear problem parameters, the hybrid method reduces the solution time by 5.7×, and reduces energy consumption by 11.6×, compared to a baseline solver running on a GPU.
|
Year | DOI | Venue |
|---|---|---|
2017 | 10.1145/3123939.3124550 | MICRO-50: The 50th Annual IEEE/ACM International Symposium on Microarchitecture
Cambridge
Massachusetts
October, 2017 |
Keywords | Field | DocType |
analog, accelerator, nonlinear, Newton's method | Graphics,Applied mathematics,Mathematical optimization,Nonlinear system,Computer science,Parallel computing,Model of computation,Solver,Analog computer,Partial differential equation,Grid,Newton's method | Conference |
ISSN | ISBN | Citations |
1072-4451 | 978-1-4503-4952-9 | 3 |
PageRank | References | Authors |
0.41 | 11 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yipeng Huang | 1 | 23 | 2.82 |
| Ning Guo | 2 | 14 | 6.52 |
| Mingoo Seok | 3 | 11 | 6.77 |
| Y. P. Tsividis | 4 | 93 | 25.89 |
| Kyle T. Mandli | 5 | 3 | 0.75 |
| Simha Sethumadhavan | 6 | 925 | 54.24 |