An Adaptive VM Provisioning Method for Large-Scale Agent-Based Traffic Simulations on the Cloud

Hanai, Masahiro; Suzumura, Toyotaro; Ventresque, Anthony; Shudo, Kazuyuki

doi:10.1109/cloudcom.2014.164

Cited by 14 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Transportation/Traffic applications range from the design of Intelligent Transportation Systems (ITS) in smart cities (Ventresque et al, 2012;Xu, Aydt, & Lees, 2012) to traffic prediction (De Grande, Boukerche, Guan, & Aljeri, 2016;Suh, Hunter, & Fujimoto, 2014). Some work has also investigated the use of cloud to support large-scale DS of transportation networks (Hanai, Suzumura, Ventresque, & Shudo, 2015;Zehe, Knoll, Cai, & Aydt, 2015).…”

Section: Distributed Simulation Applicationsmentioning

confidence: 99%

Distributed simulation: state-of-the-art and potential for operational research

Taylor

2019

European Journal of Operational Research

View full text Add to dashboard Cite

Section: Distributed Simulation Applicationsmentioning

confidence: 99%

Distributed simulation: state-of-the-art and potential for operational research

Taylor

2019

European Journal of Operational Research

View full text Add to dashboard Cite

“…In our simulation, events received from other LPs are buffered before they are inserted to the event queues (line 1). If a newly received event has a receiving time smaller than the minimum loaded time (which is initialized to infinity), then the stored events, anti-messages, and states are loaded from the storage (lines 7-9) before they are inserted to the queues (lines [14][15][16][17][18][19][20][21]. After that, the minimum loaded time is updated to the new received time (line 22), and then the newly received event is inserted to the event queue as usual (line 25).…”

Section: Implementation Of Repeating Simulationmentioning

confidence: 99%

“…Since large-scale systems consist of numerous events and states, the use of PDES to analyze them requires a large number of repeated simulations based on various what-if scenarios and parameter patterns. For example, the Tokyo traffic simulation [15,25,26,29] made 770,000 (770K) repeated simulations to analyze what happens if one of the roads was blocked, since there are 770K junctions and thus generated the same number of blocking scenarios. If pairs of blocks of the junctions are simulated, the total number of simulations exceeds 296 billion ( 770K 2 ).…”

Section: Introductionmentioning

confidence: 99%

Exact-Differential Simulation

Hanai

Suzumura

Liu

et al. 2019

ACM Trans. Model. Comput. Simul.

Self Cite

View full text Add to dashboard Cite

Using computer simulation to analyze large-scale discrete event systems requires repeated executions with various scenarios or parameters. Such repeated executions can induce significant redundancy in event processing when the modification from a prior scenario to a new scenario is relatively minor, and when the altered scenario influences only a small part of the simulation. For example, in a city-scale traffic simulation, an altered scenario of blocking one junction may only affect a small part of the city for considerable length of time. However, traditional simulation approaches would still repeat the simulation for the whole city even when the changes are minor. In this article, we propose a new redundancy reduction technique for large-scale discrete event simulations, called exact-differential simulation, which simulates only the altered portions of scenarios and their influences in repeated executions while still achieving the same results as the re-execution of entire simulations. This article presents the main concepts of the exact-differential simulation, the design of its algorithm, and an approach to build an exact-differential simulation middleware that supports multiple applications of discrete event simulation. We also evaluate our approach by using two case studies, PHOLD benchmark and a traffic simulation of Tokyo.

show abstract

“…For example in simulation of Tokyo traffic (Osogami, Imamichi, Mizuta, Morimura, Raymond, Suzumura, Takahashi, and Ide 2012, Osogami, Imamichi, Mizuta, Suzumura, and Ide 2013, Suzumura and Kanezashi 2013, Hanai, Suzumura, Ventresque, and Shudo 2014, which has 770K junctions, we need to repeat simulation tasks 770K times when we pick up one junction from the 770K junctions and simulate what happens if the junction is blocked. When we simulate multiple blocks of the junctions, we need to execute 2 770K times (the sum of combination from 770K junctions choosing 0 to 770K).…”

Section: Introductionmentioning

confidence: 99%