Abstract. The High Level Architecture promotes simulation interoperability and reusability, making it an ideal candidate to model large-scale systems. However, a large-scale simulation running in a distributed environment is often affected by the imbalance of load level at different computing hosts. Migrating processes from heavily-loaded hosts to less-loaded hosts can overcome this shortcoming. We have previously developed a SimKernel framework to execute HLA-based simulations in the Grid environment with migration support as a prominent design feature. In this paper, we will introduce a transparent migration protocol for SimKernel-based federates that minimizes migration overhead.
Network of workstations are gaining popularity against parallel and supercomputers for running large-scale simulations. The Grid enables resource sharing of computing resources at geographically distributed locations and provides the simulations with immense computing power. At the meantime, the High Level Architecture (HLA) paradigm provides a software platform and interoperability interface for simulation components to utilize distributed resources. Running HLAbased simulation over the Grid environments would shorten the simulation time, improve the hardware availability and be beneficial for the simulation community. Despite the convenience and benefits the Grid can provide, issues that hinder the adoption of Grid as the simulation hardware platform still remain. The HLA lacks the capability of resource management for running simulation in the aspect of resource discovery, federate deployment, dynamic load-balancing and fault-tolerance. Hence, this project is motivated to develop a programming framework to facilitate the HLA simulation development, with the consideration for possible integration with the Grid environment. Both dynamic load-balancing and fault-tolerance require frequent checkpointing of the simulation and saving the simulation state. Consequently, the amount of information saved will greatly impact the simulation performance. HLA simulation normally has a large amount of information to be extracted and saved, hence requiring more time and computing power for federate migration and fault recovery. It is of great importance to develop a federate model that supports easy checkpointing and has minimal state information for saving and restoration. It is also noticed that substantial effort is required for writing programs that conform to the HLA Runtime Infrastructure (RTI) specification because of its complexity. In this project, a framework with easy checkpointing capability was developed, and iii ATTENTION: The Singapore Copyright Act applies to the use of this document. Nanyang Technological University Library Master of Engineering's Thesis ABSTRACT a code library was implemented. The code library is meant for automatic code generation from user design with a Graphical User Interface (GUI), which is part of the framework. HLA Data Distribution Management is used to route simulation events (interactions) to achieve efficient use of network bandwidth. Federate migration protocols have been introduced by various research projects. However, existing protocols achieve migration by using either federation wide synchronization or a third party host, such as FTP servers, to handle the saving and restoring of migration states. Based on the framework, federate migration protocols that bypass the shortcomings identified above were developed and better migration performance is achieved. To eliminate message loss during migration process, a counter mechanism was employed. Study also shows that federate join time contributes significantly to the migration overhead. Therefore, our protocol was further modified...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.