1 / 23

Diffusion scheduling in multiagent computing system

Motivation. Architecture. Algorithms. Examples. Dynamics. Diffusion scheduling in multiagent computing system. Robert Schaefer, AGH University of Science and Technology, Kraków, Poland The Group Members: Maciej Smołka Jagiellonian University, Kraków, Poland

hunter
Download Presentation

Diffusion scheduling in multiagent computing system

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Motivation Architecture Algorithms Examples Dynamics Diffusion scheduling in multiagent computing system • Robert Schaefer, AGH University of Science and Technology, Kraków, Poland • The Group Members: • Maciej Smołka • Jagiellonian University, Kraków, Poland • Piotr Uhruski, Marek Grochowski • AGH University of Science and Technology, Kraków, Poland

  2. Motivation Motivation Architecture Algorithms Examples Dynamics Distributed computation paradigms • message passing libraries • PVM Parallel Virtual Machine (1990), MPI Message-Passing Interface (1992) • SOA (Service Oriented Architecture) • CORBA (1996), SOAP (1998) • GRID • Condor (1997), Globus (1998), OGSI/OGSA (2002) • Some drawbacks : • partially manual resources allocation • time consuming deployment and maintenance of the system • usually assuming static resources Diffusion scheduling in multiagent computing system

  3. Motivation Motivation Architecture Algorithms Examples Dynamics Distributed computing using MAS technology Application Computation + Agent logic Middleware Agents environment Heterogeneous Operating Systems Network Diffusion scheduling in multiagent computing system

  4. Motivation Architecture Algorithms Examples Dynamics Octopus Octopus Octopus CORBA CORBA CORBA Java Java Java Overview of the OCTOPUS architecture Application Sample task implementations Smart Solid Agents (scheduling, grain control) Agent SDK Middleware . . . VCN Connections Virtual Topology Diffusion scheduling in multiagent computing system

  5. Architecture Motivation Architecture Algorithms Examples Dynamics OCTOPUS Key Tasks • Execute Agents • Distributed Communication • Environment Information • Migration • Virtual Network Topology • Virtual Computation Node (VCN) • Agent’s Construction Kit Agents environment Diffusion scheduling in multiagent computing system

  6. Algorithms Motivation Architecture Algorithms Examples Dynamics Diffusion scheduling idea • Analogy to molecular diffusion phenomena • Local scheduling method – every agent is autonomously searching and allocating resources at neighbouring node • We hope to obtain the asymptotically balanced load Diffusion scheduling in multiagent computing system

  7. Motivation Architecture Algorithms Examples Dynamics Diffusion scheduling – main parameters Diffusion schduling in multiagent computing system

  8. Motivation Architecture Algorithms Examples Dynamics Diffusion scheduling algorithm Diffusion scheduling in multiagent computing system

  9. Motivation Architecture Algorithms Examples Dynamics Binding energy formulas under consideration (1) (2) Diffusion scheduling in multiagent computing system

  10. Algorithms Motivation Architecture Algorithms Examples Dynamics Controlling the computation grain – Container agent • Internal job is a dynamic structure of atomic jobs • Sequential computation of contained atomic jobs • New agent creation when the number of contained jobs exceeds the capacity of the agent Diffusion scheduling in multiagent computing system

  11. Algorithms Motivation Architecture Algorithms Examples Dynamics „Weak” synchronization strategy – „Leo the Professional” agent (J. Momot, K. Kossacki – 2004) • Migrates through the network and gathers information about computing agents • Responsible for removing redundancy • Allows to avoid total synchronization of the local system Diffusion scheduling in multiagent computing system

  12. Tests Motivation Architecture Algorithms Examples Dynamics Speedup vs. grain in CAE computation Diffusion scheduling in multiagent computing system

  13. Tests Motivation Architecture Algorithms Examples Dynamics Overhead of the Agent Oriented technology (the case of HGS computation) Diffusion scheduling in multiagent computing system

  14. Tests Motivation Architecture Algorithms Examples Dynamics Speedup of the Diffusion Scheduling (the case of HGS computation) Diffusion scheduling in multiagent computing system

  15. Motivation Architecture Algorithms Examples Dynamics Communication dependent rules „WAN” emulation „LAN” case Diffusion scheduling in multiagent computing system

  16. Motivation Architecture Algorithms Examples Dynamics Experiments in the local area network (1) (2) Diffusion scheduling in multiagent computing system

  17. Motivation Architecture Algorithms Examples Dynamics Experiments in the wide area network (1) (2) Diffusion scheduling in multiagent computing system

  18. Conclusions Motivation Architecture Algorithms Examples Dynamics Preliminaries Diffusion scheduling in multiagent computing system

  19. Conclusions Motivation Architecture Algorithms Examples Dynamics State equations Diffusion scheduling in multiagent computing system

  20. Conclusions Motivation Architecture Algorithms Examples Dynamics Optimal scheduling problem Diffusion scheduling in multiagent computing system

  21. Conclusions • Diffusion scheduling is an effective tool of managing large-scale distributed systems. It is achieved by the low complexity of local scheduling rules and only local communication. It ensures proper agent location in the dynamic network environment. • Introduced formal description provides the discrete equation of evolution and the characterization of admissible controls as well as the cost functional for computing MAS. • The optimal scheduling problem posses the unique solution in the class of stationary strategies. • Total overhead is low in comparison with the computation time (~ 5%). • No significant requirements imposed over applications. Diffusion scheduling in multiagent computing system

  22. Thank you for your patience! Diffusion scheduling in multiagent computing system

  23. Publications

More Related