T. Lux, L. Watson, Tyler H. Chang, Jon Bernard, Bo Li, Li Xu, Godmar Back, A. Butt, K. Cameron, Yili Hong
{"title":"Predictive modeling of I/O characteristics in high performance computing systems","authors":"T. Lux, L. Watson, Tyler H. Chang, Jon Bernard, Bo Li, Li Xu, Godmar Back, A. Butt, K. Cameron, Yili Hong","doi":"10.22360/springsim.2018.hpc.009","DOIUrl":"https://doi.org/10.22360/springsim.2018.hpc.009","url":null,"abstract":"Each of high performance computing, cloud computing, and computer security have their own interests in modeling and predicting the performance of computers with respect to how they are configured. An effective model might infer internal mechanics, minimize power consumption, or maximize computational throughput of a given system. This paper analyzes a four-dimensional dataset measuring the input/output (I/O) characteristics of a cluster of identical computers using the benchmark IOzone. The I/O performance characteristics are modeled with respect to system configuration using multivariate interpolation and approximation techniques. The analysis reveals that accurate models of I/O characteristics for a computer system may be created from a small fraction of possible configurations, and that some modeling techniques will continue to perform well as the number of system parameters being modeled increases. These results have strong implications for future predictive analyses based on more comprehensive sets of system parameters.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"9 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116079598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Simulating link aggregation in private virtual lan using openflow for cloud environment","authors":"Damilola Murtala, Yasir Malik, P. Zavarsky","doi":"10.22360/springsim.2018.cns.014","DOIUrl":"https://doi.org/10.22360/springsim.2018.cns.014","url":null,"abstract":"Segregation and isolation of mission critical devices and services are among the main security concerns in cloud computing environments. Private Virtual LAN (PVLAN) offers the ability to efficiently support segregation and isolation among end devices. Link aggregation on PVLAN promiscuous ports reduces the risk of single point of failure for the entire PVLAN network. This research focuses on improving security and availability of nodes within the PVLAN domain and layer three devices by combining multiple PVLAN promiscuous ports as a single logical port using Software Defined Networking protocols. Our approach enables cloud platform to implement PVLAN by incorporating link aggregation to extend and support PVLAN features for optimal load balancing and path selection of inbound and outbound traffic. It also helps to reduce network inefficiencies which might occur from multiple traffic utilizing a single communication uplink. Simulation results show the effectiveness of our approach.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"31 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116650449","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Tolk, F. Barros, A. D’Ambrogio, A. Rajhans, P. Mosterman, S. Shetty, M. Traoré, H. Vangheluwe, L. Yilmaz
{"title":"Hybrid simulation for cyber physical systems: a panel on where are we going regarding complexity, intelligence, and adaptability of CPS using simulation","authors":"A. Tolk, F. Barros, A. D’Ambrogio, A. Rajhans, P. Mosterman, S. Shetty, M. Traoré, H. Vangheluwe, L. Yilmaz","doi":"10.22360/springsim.2018.msciaas.004","DOIUrl":"https://doi.org/10.22360/springsim.2018.msciaas.004","url":null,"abstract":"During the Spring Simulation Multi-Conference 2017, a group of invited experts discussed challenges in M&S of cyber physical systems. This 2018 panel is a follow-on activity, asking how the combination of various simulation paradigms, methods - so-called hybrid simulation - can be utilized regarding complexity, intelligence, and adaptability of cyber physical systems. This paper is a collection of position papers of the participating experts, supporting their viewpoints represented in the discussion.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127801935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The achievement gap phenomenon: a practical application of modeling & simulation","authors":"IV JohnJ.Johnson, J. Padilla, S. Diallo","doi":"10.22360/springsim.2018.anss.024","DOIUrl":"https://doi.org/10.22360/springsim.2018.anss.024","url":null,"abstract":"The academic achievement gap is a persistent phenomenon in U.S. education system despite a long history of efforts and billions of dollars spent to correct it. Literature abounds with theories about why the gap exist, such as: student self-perception, parent involvement, teacher quality, and others. Model based approaches have been used to understand various aspects of the phenomenon. However, no models were identified that consider a comprehensive set of theories, and is specifically designed to investigate potential policies and strategies for reducing the gap. We build such a model using a methodology that includes: a) Modeling and Simulation-System Development Framework (MS-SDF); b) Systems Modeling Language (SysML); and c) a Systems Dynamics approach. Preliminary findings indicate that concepts from prevailing theories about the achievement gap can be accurately represented in a single system dynamics model. We also identify key stakeholders, functions, and variables affecting the achievement gap.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"535 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124517726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"IoT, smart homes, and zigbee simulation","authors":"H. Rajaei, F. Mirzaei","doi":"10.22360/springsim.2018.cns.009","DOIUrl":"https://doi.org/10.22360/springsim.2018.cns.009","url":null,"abstract":"Internet of Things (IoT) is predicted to connect more than 50 billion devices to each other via the Internet by 2020. Smart homes and buildings are primary targets to continuously monitor the connected devices, collect data, process them, and provide feedback where and when it is needed. In this chain process of IoT, there are numerous nontrivial communications and computational issues. In this paper, we focus on smart homes and specifically on ZigBee, as a smart data collection element. We view several topologies and configurations using ZigBee connectivity in a typical smart home. In addition, we report a simulation study to shed some light on how ZigBee networks perform. To transfer the huge amount of collected data from smart homes in a timely manner and process it, we use the Fifth Generation (5G) of wireless network, and for processing the data we use Mobile Edge Computing (MEC) interacted with cloud.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114005420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Using simulation to measure the performance of a search operation for a man overboard","authors":"M. Karatas, B. Onggo, M. Gunal","doi":"10.22360/springsim.2018.anss.009","DOIUrl":"https://doi.org/10.22360/springsim.2018.anss.009","url":null,"abstract":"A man overboard is defined as a situation in which a person falls out at sea from a ship and is in need of rescue. In such cases, a quick and effective search and rescue (SAR) operation are crucial to increase the survival probability of the victim. If there is no visual contact with the victim, i.e. the exact time and location that the crew was overboard, and the drift direction and speed in the region are not certain, then the situation requires conducting a rescue maneuver in an uncertain area that is expanding over time. In this study, we compute the lower and upper bounds on the performance of a search operation in an expanding area for a man overboard situation. After deriving bounds on the cumulative detection probability of the victim, we test the validity of our analytic results by performing Monte Carlo simulations.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122041108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Introduction to parallel DEVS modelling and simulation","authors":"Yentl Van Tendeloo, H. Vangheluwe","doi":"10.22360/springsim.2018.mod4sim.012","DOIUrl":"https://doi.org/10.22360/springsim.2018.mod4sim.012","url":null,"abstract":"DEVS is a popular formalism for modelling complex dynamic systems using a discrete-event abstraction. Main advantages of DEVS are its rigorous formal definition, and its support for modularity: models can be hierarchically nested. Thanks to these properties, DEVS frequently serves as a simulation \"assembly language\" to which models in other formalisms are mapped. This makes it possible to combine models in different formalisms together by mapping both to DEVS. This tutorial introduces the practical use of the Parallel DEVS formalism in a bottom-up fashion. We start from simple autonomous Atomic (i.e., non-hierarchical) DEVS models and increment up to Coupled (i.e., hierarchical) DEVS models. Each increment is illustrated with a minimal running example. The focus is on the practical use of DEVS modelling and simulation, though necessary theoretical foundations are interleaved. Examples are presented using Python-PDEVS, though the foundations and techniques apply to other DEVS simulation tools as well.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115733327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Design patterns for variability modeling using SES ontology","authors":"C. Deatcu, H. Folkerts, T. Pawletta, Umut Durak","doi":"10.22360/springsim.2018.mod4sim.004","DOIUrl":"https://doi.org/10.22360/springsim.2018.mod4sim.004","url":null,"abstract":"The System Entity Structure (SES) is a high level approach for variability modeling, particularly in simulation engineering, which is under continuous development. In this context, an enhanced framework is introduced that supports dynamic variability evolution using the SES approach. However, the main focus is to start a discussion about a set of design patterns, which were developed to analyze the tree design and computing aspects of System Entity Structures. As development of our MATLAB-based SES toolbox for construction and pruning of SES trees proceeded, the necessity to have some generalized examples for testing and verification came more and more into awareness. We propose a set of design patterns that, if completely representable and computable by a certain tool, support all aspects of SES theory. In addition, the patterns give users substantial support for developing SES models for other applications.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115961001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alexander Wichmann, Ralph Maschotta, Francesco Bedini, A. Zimmermann
{"title":"A workflow for the design of optimized system architectures using model-driven optimization","authors":"Alexander Wichmann, Ralph Maschotta, Francesco Bedini, A. Zimmermann","doi":"10.22360/springsim.2018.mod4sim.002","DOIUrl":"https://doi.org/10.22360/springsim.2018.mod4sim.002","url":null,"abstract":"Model-based design of complex systems benefits from the early validation of design decisions. A direct evaluation is usually not possible because the systems are too complicated. Thus, indirect optimization with simulation-based validation is the method of choice. An approach for model-driven optimization of system architectures based on the UML standard has been proposed in our earlier work and implemented in a software framework. To close the gap between standardized UML system model and domain-specific models for the evaluation tool, model transformations between both domains are required. This paper describes the workflow for model-based design of complex systems using a model-driven optimization framework.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"53 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127336489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. Aymerich, Alexandre Duchateau, E. Montagne, Frank Plochan
{"title":"On tuning the symmetric sparse matrix vector multiplication with CSR and TJDS","authors":"E. Aymerich, Alexandre Duchateau, E. Montagne, Frank Plochan","doi":"10.22360/springsim.2018.hpc.005","DOIUrl":"https://doi.org/10.22360/springsim.2018.hpc.005","url":null,"abstract":"In this work we present a heuristic to select the appropriate compressed storage format when computing the symmetric SpMV multiplication sequentially. A subset of symmetric sparse matrices were selected from the SPARSITY benchmark suite and extended with other matrices we consider complement them. All matrices were collected from Matrix Market and UF matrix collection. Experimental evidence shows that given a symmetric sparse matrix, predicting what is the more convenient format to use for computing the symmetric SpMV multiplication could be possible. According to our findings, and good rule of thumb, if the average number of non zero coefficients per column (row) is less than 3.5, then the symmetric SpMV multiplication runs up to 1.6x faster using the TJDS format compared to CSR.","PeriodicalId":413389,"journal":{"name":"Spring Simulation Multiconference","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134571785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}