Distributed Java Programs Initial Mapping ...
Document type :
Communication dans un congrès avec actes
Title :
Distributed Java Programs Initial Mapping Based on Extremal Optimization
Author(s) :
Olejnik, Richard [Auteur]
LIFL - DART/Émeraude
Contributions of the Data parallelism to real time [DART]
Laboratoire d'Informatique Fondamentale de Lille [LIFL]
de Falco, Ivanoe [Auteur]
Laskowski, Eryk [Auteur]
Institute of Computer Science [Warszawa]
Scafuri, Umberto [Auteur]
Tarantino, Ernesto [Auteur]
Institute of High Performance Computing and Networking [ICAR]
Tudruj, Marek [Auteur]
Polish-Japanese Institute of Information Technology [PJIIT]
Institute of Computer Science [Warszawa]

LIFL - DART/Émeraude
Contributions of the Data parallelism to real time [DART]
Laboratoire d'Informatique Fondamentale de Lille [LIFL]
de Falco, Ivanoe [Auteur]
Laskowski, Eryk [Auteur]
Institute of Computer Science [Warszawa]
Scafuri, Umberto [Auteur]
Tarantino, Ernesto [Auteur]
Institute of High Performance Computing and Networking [ICAR]
Tudruj, Marek [Auteur]
Polish-Japanese Institute of Information Technology [PJIIT]
Institute of Computer Science [Warszawa]
Scientific editor(s) :
Springer-Verlag Berlin Heidelberg
Conference title :
Applied Parallel and Scientific Computing - 10th International Conference, PARA 2010 (EUROPAR)- Revised Selected Papers, Part I
City :
Reykjavik
Country :
Islande
Start date of the conference :
2010-06-06
Book title :
PARA (1)
Journal title :
Lecture Notes in Computer Science
Publisher :
Springer-Verlag Berlin Heidelberg
Publication date :
2012
English keyword(s) :
distributed systems
program optimization
evolutionary algorithm
program optimization
evolutionary algorithm
HAL domain(s) :
Informatique [cs]/Calcul parallèle, distribué et partagé [cs.DC]
English abstract : [en]
An extremal optimization algorithm for initial Java program placement on clusters of Java Virtual Machines (JVMs) is presented. JVMs are implemented on multicore processors working under the ProActive Java execution ...
Show more >An extremal optimization algorithm for initial Java program placement on clusters of Java Virtual Machines (JVMs) is presented. JVMs are implemented on multicore processors working under the ProActive Java execution framework. Java programs are represented as Directed Acyclic Graphs in which tasks correspond to methods of distributed active Java objects that communicate using a RMI mechanism. The presented probabilistic extremal optimization approach is based on the local fitness function composed of two sub-functions in which elimination of delays of task execution after reception of required data and the imbalance of tasks execution in processors are used as heuristics for improvements of extremal optimization solutions. The evolution of an extremal optimization solution is governed by task clustering supported by identification of the dominant path in the graph. The applied task mapping is based on dynamic measurements of current loads of JVMs and inter-JVM communication link bandwidth. The JVM loads are approximated by observation of the average idle time that threads report to the OS. The current link bandwidth is determined by observation of the performed average number of RMI calls per second.Show less >
Show more >An extremal optimization algorithm for initial Java program placement on clusters of Java Virtual Machines (JVMs) is presented. JVMs are implemented on multicore processors working under the ProActive Java execution framework. Java programs are represented as Directed Acyclic Graphs in which tasks correspond to methods of distributed active Java objects that communicate using a RMI mechanism. The presented probabilistic extremal optimization approach is based on the local fitness function composed of two sub-functions in which elimination of delays of task execution after reception of required data and the imbalance of tasks execution in processors are used as heuristics for improvements of extremal optimization solutions. The evolution of an extremal optimization solution is governed by task clustering supported by identification of the dominant path in the graph. The applied task mapping is based on dynamic measurements of current loads of JVMs and inter-JVM communication link bandwidth. The JVM loads are approximated by observation of the average idle time that threads report to the OS. The current link bandwidth is determined by observation of the performed average number of RMI calls per second.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Comment :
11 pages
Collections :
Source :
Files
- https://api.istex.fr/ark:/67375/HCB-33X342HN-5/fulltext.pdf?sid=hal
- Open access
- Access the document
- https://api.istex.fr/ark:/67375/HCB-33X342HN-5/fulltext.pdf?sid=hal
- Open access
- Access the document
- https://api.istex.fr/ark:/67375/HCB-33X342HN-5/fulltext.pdf?sid=hal
- Open access
- Access the document
- https://api.istex.fr/ark:/67375/HCB-33X342HN-5/fulltext.pdf?sid=hal
- Open access
- Access the document
- fulltext.pdf
- Open access
- Access the document