Predetermined Rollbacks: An extension to Time Warp for spatially parallel agent-based simulation

Date

2019-09-01

Author

Gorur, Bilge Kaan
İMRE, KAYHAN MUSTAFA
Oğuztüzün, Mehmet Halit S.
Yilmaz, Levent

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

54
views

0
downloads

Time management is an important factor that affects the speed of parallel and distributed simulations. Conservative time management mechanisms advance simulation time after synchronization of logical processes, whereas optimistic time management mechanisms loosen synchronization among processes to speed up simulation. Because of loosened synchronization, the optimistic approaches are vulnerable to causality faults that must be addressed by logical processes during run-time. Repairing a simulation, by means of a rollback mechanism, is one of the most time-consuming processes of optimistic approaches. In this paper, we propose a method that considers precautionary measures against possible future rollbacks for Time Warp mechanism due to Jefferson. Our proposed method, named Predetermined Rollbacks, uses a modified simulation engine that can detect and avoid unnecessary rollbacks. Our experiments demonstrate that the proposed method can significantly improve the speedup of Time Warp in agent-based simulations, where agents communicate in a shared environment.

Subject Keywords

Modelling and Simulation, Hardware and Architecture, Software

URI

https://hdl.handle.net/11511/42898

Journal

SIMULATION MODELLING PRACTICE AND THEORY

DOI

https://doi.org/10.1016/j.simpat.2019.04.008

Collections

Department of Computer Engineering, Article

Suggestions

OpenMETU
Core

On the use of model-driven engineering principles for the management of simulation experiments Dayibas, Orcun; Oğuztüzün, Mehmet Halit S.; Yilmaz, Levent (Informa UK Limited, 2019-04-03) Simulation experiments are an essential part of computational science and engineering. The use of simulation models is widely adopted by practitioners from diverse areas of applied sciences. Nevertheless, simulations are rarely replicated due to reuse and maintenance challenges related to models and data. In this respect, we propose that crucial and labor intensive parts of simulation experiments could be supported by model transformations. This work focuses on model-driven engineering practices to enable r...
Modeling cultures of the embedded software industry: feedback from the field Akdur, Deniz; Say, Bilge; Demirörs, Onur (Springer Science and Business Media LLC, 2020-06-01) Engineering of modern embedded systems requires complex technical, managerial and operational processes. To cope with the complexity, modeling is a commonly used approach in the embedded software industry. The modeling approaches in embedded software vary since the characteristics of modeling such as purpose, medium type and life cycle phase differ among systems and industrial sectors. The objective of this paper is to detail the use of a characterization model MAPforES ("Modeling Approach Patterns for Embe...
A framework for reviewing domain specific conceptual models TANRIÖVER, ÖMER ÖZGÜR; Bilgen, Semih (Elsevier BV, 2011-09-01) Conceptual models are used in understanding and communicating the domain of interest during analysis phase of system development. As they are used in early phases, errors and omissions may propagate to later phases and may be very costly to correct. This paper proposes a framework for evaluating conceptual models when represented in a domain specific language based on UML constructs. The framework describes the main aspects to be considered when conceptual models are represented in a domain specific languag...
Machine learning algorithms for accurate flow-based network traffic classification: Evaluation and comparison Soysal, Murat; Schmidt, Şenan Ece (Elsevier BV, 2010-06-01) The task of network management and monitoring relies on an accurate characterization of network traffic generated by different applications and network protocols. We employ three supervised machine learning (ML) algorithms, Bayesian Networks, Decision Trees and Multilayer Perceptrons for the flow-based classification of six different types of Internet traffic including peer-to-peer (P2P) and content delivery (Akamai) traffic. The dependency of the traffic classification performance on the amount and composi...
Schedulability analysis of real-time multi-frame co-simulations on multi-core platforms Ahsan, Muhammad Uzair; Oğuztüzün, Mehmet Halit S.; Department of Computer Engineering (2020) For real-time simulations, the fidelity of simulation does not depend only on the functional accuracy of simulation but also on its timeliness. It is helpful for developers if we can analyze and verify that a simulation will always meet its timing requirements while keeping an acceptable level of accuracy. Abstracting the simulated processes simply as software tasks allows us to transform the problem of verifying timeliness into a schedulability analysis problem where tasks are checked if they are schedulab...

Citation Formats

B. K. Gorur, K. M. İMRE, M. H. S. Oğuztüzün, and L. Yilmaz, “Predetermined Rollbacks: An extension to Time Warp for spatially parallel agent-based simulation,” SIMULATION MODELLING PRACTICE AND THEORY, pp. 60–77, 2019, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/42898.