Distributed Java Programs Initial Mapping Based on Extremal Optimization
11 pages International audience 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...
Main Authors: | , , , , , |
---|---|
Other Authors: | , , , , , , , , , , , , , , , , |
Format: | Conference Object |
Language: | English |
Published: |
HAL CCSD
2010
|
Subjects: | |
Online Access: | https://hal.science/hal-00834016 https://doi.org/10.1007/978-3-642-28151-8_8 |
id |
ftccsdartic:oai:HAL:hal-00834016v1 |
---|---|
record_format |
openpolar |
institution |
Open Polar |
collection |
Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe) |
op_collection_id |
ftccsdartic |
language |
English |
topic |
distributed systems program optimization evolutionary algorithm LNCS [INFO.INFO-DC]Computer Science [cs]/Distributed Parallel and Cluster Computing [cs.DC] |
spellingShingle |
distributed systems program optimization evolutionary algorithm LNCS [INFO.INFO-DC]Computer Science [cs]/Distributed Parallel and Cluster Computing [cs.DC] Olejnik, Richard de Falco, Ivanoe Laskowski, Eryk Scafuri, Umberto Tarantino, Ernesto Tudruj, Marek Distributed Java Programs Initial Mapping Based on Extremal Optimization |
topic_facet |
distributed systems program optimization evolutionary algorithm LNCS [INFO.INFO-DC]Computer Science [cs]/Distributed Parallel and Cluster Computing [cs.DC] |
description |
11 pages International audience 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. |
author2 |
Laboratoire d'Informatique Fondamentale de Lille (LIFL) Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS) Contributions of the Data parallelism to real time (DART) Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Inria Lille - Nord Europe Institut National de Recherche en Informatique et en Automatique (Inria) LIFL - DART/Émeraude Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS) Institute of High Performance Computing and Networking, CNR, Naples, Italy Institute of High Performance Computing and Networking (ICAR) National Research Council of Italy Institute of Computer Science Warszawa Polska Akademia Nauk = Polish Academy of Sciences = Académie polonaise des sciences (PAN) Polish-Japanese Institute of Information Technology (PJIIT) PICS CNRS Springer-Verlag Berlin Heidelberg Académie des Sciences de Pologne CNRS Italie |
format |
Conference Object |
author |
Olejnik, Richard de Falco, Ivanoe Laskowski, Eryk Scafuri, Umberto Tarantino, Ernesto Tudruj, Marek |
author_facet |
Olejnik, Richard de Falco, Ivanoe Laskowski, Eryk Scafuri, Umberto Tarantino, Ernesto Tudruj, Marek |
author_sort |
Olejnik, Richard |
title |
Distributed Java Programs Initial Mapping Based on Extremal Optimization |
title_short |
Distributed Java Programs Initial Mapping Based on Extremal Optimization |
title_full |
Distributed Java Programs Initial Mapping Based on Extremal Optimization |
title_fullStr |
Distributed Java Programs Initial Mapping Based on Extremal Optimization |
title_full_unstemmed |
Distributed Java Programs Initial Mapping Based on Extremal Optimization |
title_sort |
distributed java programs initial mapping based on extremal optimization |
publisher |
HAL CCSD |
publishDate |
2010 |
url |
https://hal.science/hal-00834016 https://doi.org/10.1007/978-3-642-28151-8_8 |
op_coverage |
Reykjavik, Iceland |
genre |
Iceland |
genre_facet |
Iceland |
op_source |
ISSN: 0302-9743 Lecture Notes in Computer Science PARA (1) Applied Parallel and Scientific Computing - 10th International Conference, PARA 2010 (EUROPAR)- Revised Selected Papers, Part I https://hal.science/hal-00834016 Applied Parallel and Scientific Computing - 10th International Conference, PARA 2010 (EUROPAR)- Revised Selected Papers, Part I, Jun 2010, Reykjavik, Iceland. pp. 180-191, ⟨10.1007/978-3-642-28151-8_8⟩ |
op_relation |
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-642-28151-8_8 hal-00834016 https://hal.science/hal-00834016 doi:10.1007/978-3-642-28151-8_8 |
op_doi |
https://doi.org/10.1007/978-3-642-28151-8_8 |
container_start_page |
75 |
op_container_end_page |
85 |
_version_ |
1789967496663531520 |
spelling |
ftccsdartic:oai:HAL:hal-00834016v1 2024-02-04T10:01:32+01:00 Distributed Java Programs Initial Mapping Based on Extremal Optimization Olejnik, Richard de Falco, Ivanoe Laskowski, Eryk Scafuri, Umberto Tarantino, Ernesto Tudruj, Marek Laboratoire d'Informatique Fondamentale de Lille (LIFL) Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS) Contributions of the Data parallelism to real time (DART) Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Inria Lille - Nord Europe Institut National de Recherche en Informatique et en Automatique (Inria) LIFL - DART/Émeraude Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS)-Université de Lille, Sciences et Technologies-Institut National de Recherche en Informatique et en Automatique (Inria)-Université de Lille, Sciences Humaines et Sociales-Centre National de la Recherche Scientifique (CNRS) Institute of High Performance Computing and Networking, CNR, Naples, Italy Institute of High Performance Computing and Networking (ICAR) National Research Council of Italy Institute of Computer Science Warszawa Polska Akademia Nauk = Polish Academy of Sciences = Académie polonaise des sciences (PAN) Polish-Japanese Institute of Information Technology (PJIIT) PICS CNRS Springer-Verlag Berlin Heidelberg Académie des Sciences de Pologne CNRS Italie Reykjavik, Iceland 2010-06-06 https://hal.science/hal-00834016 https://doi.org/10.1007/978-3-642-28151-8_8 en eng HAL CCSD Springer-Verlag Berlin Heidelberg Springer info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-642-28151-8_8 hal-00834016 https://hal.science/hal-00834016 doi:10.1007/978-3-642-28151-8_8 ISSN: 0302-9743 Lecture Notes in Computer Science PARA (1) Applied Parallel and Scientific Computing - 10th International Conference, PARA 2010 (EUROPAR)- Revised Selected Papers, Part I https://hal.science/hal-00834016 Applied Parallel and Scientific Computing - 10th International Conference, PARA 2010 (EUROPAR)- Revised Selected Papers, Part I, Jun 2010, Reykjavik, Iceland. pp. 180-191, ⟨10.1007/978-3-642-28151-8_8⟩ distributed systems program optimization evolutionary algorithm LNCS [INFO.INFO-DC]Computer Science [cs]/Distributed Parallel and Cluster Computing [cs.DC] info:eu-repo/semantics/conferenceObject Conference papers 2010 ftccsdartic https://doi.org/10.1007/978-3-642-28151-8_8 2024-01-06T23:22:25Z 11 pages International audience 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. Conference Object Iceland Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe) 75 85 |