Zobrazeno 1 - 8
of 8
pro vyhledávání: '"Lockhart, Shelby"'
Supercomputer architectures are trending toward higher computational throughput due to the inclusion of heterogeneous compute nodes. These multi-GPU nodes increase on-node computational efficiency, while also increasing the amount of data to be commu
Externí odkaz:
http://arxiv.org/abs/2209.06141
Krylov methods are a key way of solving large sparse linear systems of equations, but suffer from poor strong scalabilty on distributed memory machines. This is due to high synchronization costs from large numbers of collective communication calls al
Externí odkaz:
http://arxiv.org/abs/2203.06144
Publikováno v:
Proceedings of the 2022 SIAM Conference on Parallel Processing for Scientific Computing
Anderson Acceleration (AA) is a method to accelerate the convergence of fixed point iterations for nonlinear, algebraic systems of equations. Due to the requirement of solving a least squares problem at each iteration and a reliance on modified Gram-
Externí odkaz:
http://arxiv.org/abs/2110.09667
The cost of data movement on parallel systems varies greatly with machine architecture, job partition, and nearby jobs. Performance models that accurately capture the cost of data movement provide a tool for analysis, allowing for communication bottl
Externí odkaz:
http://arxiv.org/abs/2010.10378
Publikováno v:
In Parallel Computing July 2023 116
In this paper, we solve the l2-l1 sparse recovery problem by transforming the objective function of this problem into an unconstrained differentiable function and apply a limited-memory trust-region method. Unlike gradient projection-type methods, wh
Externí odkaz:
http://arxiv.org/abs/1602.08813
Publikováno v:
Proceedings of SPIE; 9/27/2017, Vol. 10394, p1-8, 8p
Autor:
Lu, Yue M., Van De Ville, Dimitri, Papadakis, Manos, Adhikari, Lasith, DeGuchy, Omar, Erway, Jennifer B., Lockhart, Shelby, Marcia, Roummel F.
Publikováno v:
Proceedings of SPIE; August 2017, Vol. 10394 Issue: 1 p103940J-103940J-8, 935469p