|
Title:
|
LEARNING LOAD BALANCING FOR SIMULATION IN HETEROGENEOUS SYSTEMS |
|
Author(s):
|
Irina Bernst, Christof Kaufmann, Jörg Frochte |
|
ISBN:
|
978-989-8533-45-6 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2015 |
|
Edition:
|
Single |
|
Keywords:
|
Machine Learning, Data Mining, Load Balancing, Assistant System, Distributed Computing, FEM |
|
Type:
|
Full Paper |
|
First Page:
|
121 |
|
Last Page:
|
128 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Distributed computing is an important key technology for simulation technologies like finite elements. Dedicated homogeneous parallel servers are well-scaling but very expensive solutions, while heterogeneous systems are a challenge for efficient computing. The work we present deals with the problem to provide a learning assistance system for load balancing in FEM simulations. Our method uses a two-stage architecture to minimize additional computational costs. The approach does neither require the labeled data nor a teacher for the initial setup and can improve itself unsupervised. For the FEM application case we introduce a novel feature set and perform an evaluation for several problem sets. |
|
|
|
|
|
|
