|
Title:
|
AN APPROACH FOR LOAD BALANCING FOR SIMULATION IN HETEROGENEOUS DISTRIBUTED SYSTEMS USING SIMULATION DATA MINING |
|
Author(s):
|
Irina Bernst, Patrick Bouillon, Jörg Frochte, Christof Kaufmann |
|
ISBN:
|
978-989-8533-25-8 |
|
Editors:
|
Hans Weghorn |
|
Year:
|
2014 |
|
Edition:
|
Single |
|
Keywords:
|
Machine learning; Simulation Data Mining; Load balancing; Distributed Systems; Finite Element Method. |
|
Type:
|
Short Paper |
|
First Page:
|
254 |
|
Last Page:
|
258 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This paper describes an approach to reduce the computation time of finite element simulations on heterogeneous distributed systems. This should be achieved by enhanced load balancing with help of machine learning techniques. Based on the hardware topology and the finite element problem the machine learning algorithm would be trained to predict the computation time in dependence on the geometric partitioning. The learned model will then be optimized to find the best partitioning regarding the computation time. The challenge of load balancing on non-homogeneous clusters is to be solved to make distributed computing an accepted method for industrial users in the field of simulations. |
|
|
|
|
|
|
