Unambiguous experiment descriptions are increasingly required for model publication, as they contain information important for reproducing simulation results. In the context of model composition, this information can be used to generate experiments for the composed model. If the original experiment descriptions specify which model property they refer to, we can then execute the generated experiments and assess the validity of the composed model by evaluating their results. Thereby, we move the attention to describing properties of a model's behavior and the conditions under which these hold, i.e., its semantics. We illuminate the potential of this concept by considering the composition of Lotka-Volterra models. In a first prototype realized for JAMES II, we use ML-Rules to describe and execute the Lotka-Volterra models and SESSL for specifying the original experiments. Model properties are described in continuous stochastic logic, and we use statistical model checking for their evaluation. Based on this, experiments to check whether these properties hold for the composed model are automatically generated and executed.
{"title":"Towards semantic model composition via experiments","authors":"Danhua Peng, Roland Ewald, A. Uhrmacher","doi":"10.1145/2601381.2601394","DOIUrl":"https://doi.org/10.1145/2601381.2601394","url":null,"abstract":"Unambiguous experiment descriptions are increasingly required for model publication, as they contain information important for reproducing simulation results. In the context of model composition, this information can be used to generate experiments for the composed model. If the original experiment descriptions specify which model property they refer to, we can then execute the generated experiments and assess the validity of the composed model by evaluating their results. Thereby, we move the attention to describing properties of a model's behavior and the conditions under which these hold, i.e., its semantics. We illuminate the potential of this concept by considering the composition of Lotka-Volterra models. In a first prototype realized for JAMES II, we use ML-Rules to describe and execute the Lotka-Volterra models and SESSL for specifying the original experiments. Model properties are described in continuous stochastic logic, and we use statistical model checking for their evaluation. Based on this, experiments to check whether these properties hold for the composed model are automatically generated and executed.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122654297","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}
Jared Ivey School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 j.ivey@gatech.edu George Riley School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 riley@ece.gatech.edu Brian Swenson School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 bswenson3@gatech.edu
{"title":"Phold performance for distributed network simulation under conservative synchronization methods in ns-3","authors":"Jared S. Ivey, G. Riley, B. Swenson","doi":"10.1145/2601381.2601410","DOIUrl":"https://doi.org/10.1145/2601381.2601410","url":null,"abstract":"Jared Ivey School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 j.ivey@gatech.edu George Riley School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 riley@ece.gatech.edu Brian Swenson School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA 30332-0250 bswenson3@gatech.edu","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121896641","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}
Performance is one of the major concerns in large-scale parallel microscopic traffic simulations. This paper focuses on one of the most time-costly data structures: the two-dimensional spatial index. A drawback of using popular two-dimensional tree-based spatial indexes (e.g. the R*-Tree) in large-scale microscopic traffic simulation is the heavy cost to rebalance the tree structure when a large number of vehicles frequently update their locations. This heavy location update cost also reduces the scalability of parallel microscopic traffic simulations. We observe that in real-world traffic systems the road density during a short period is stable, which is not sensitive to an individual vehicle's location. Thus, why not build a balanced tree structure based on the average road density in a road network? Motivated by this observation, this paper proposes Sim-Tree. The key feature of the Sim-Tree is that there is no need to check or rebalance its tree structure when individual vehicles frequently update their locations. In addition, a rebalance function and a bottom-up region query function are designed to optimize Sim-Tree's region query operations. The results of experiments simulating a city-scale traffic scenario on a 6-core machine show that the Sim-Tree is scalable and performs significantly better than the R*-tree family of spatial indexes.
{"title":"Sim-Tree: indexing moving objects in large-scale parallel microscopic traffic simulation","authors":"Yan Xu, Gary S. H. Tan","doi":"10.1145/2601381.2601388","DOIUrl":"https://doi.org/10.1145/2601381.2601388","url":null,"abstract":"Performance is one of the major concerns in large-scale parallel microscopic traffic simulations. This paper focuses on one of the most time-costly data structures: the two-dimensional spatial index. A drawback of using popular two-dimensional tree-based spatial indexes (e.g. the R*-Tree) in large-scale microscopic traffic simulation is the heavy cost to rebalance the tree structure when a large number of vehicles frequently update their locations. This heavy location update cost also reduces the scalability of parallel microscopic traffic simulations. We observe that in real-world traffic systems the road density during a short period is stable, which is not sensitive to an individual vehicle's location. Thus, why not build a balanced tree structure based on the average road density in a road network? Motivated by this observation, this paper proposes Sim-Tree. The key feature of the Sim-Tree is that there is no need to check or rebalance its tree structure when individual vehicles frequently update their locations. In addition, a rebalance function and a bottom-up region query function are designed to optimize Sim-Tree's region query operations. The results of experiments simulating a city-scale traffic scenario on a 6-core machine show that the Sim-Tree is scalable and performs significantly better than the R*-tree family of spatial indexes.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"2 9","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132581657","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}
Automatic parallelization of models has been the "Holy Grail" of the PDES community for the last 20 years. In this paper we present LORAIN -- Low Overhead Runtime Assisted Instruction Negation -- a tool capable of automatic emission of a reverse event handler by the compiler. Upon detection of certain instructions, LORAIN is able to account for, and in many cases reverse, the computation without resorting to state-saving techniques. For our PDES framework, we use Rensselaer's Optimistic Simulation System (ROSS) coupled with the LLVM compiler to generate the reverse event handler. One of the primary contributions of this work is that LORAIN operates on the LLVM-generated Intermediate Representation (IR) as opposed to the model, high-level source code. Through information gleaned from the IR, LORAIN is able to analyze, instrument, and invert various operations and emit efficient reverse event handlers at the binary code level. This preliminary work demonstrates the potential of this tool. We are able to reverse both the PHOLD model (a synthetic benchmark) as well as Fujimoto's airport model. Our results demonstrate that LORAIN-generated models are able to execute at a rate that is over 97% of hand-written, parallel model code performance.
{"title":"LORAIN: a step closer to the PDES 'holy grail'","authors":"Justin M. LaPre, Elsa Gonsiorowski, C. Carothers","doi":"10.1145/2601381.2601397","DOIUrl":"https://doi.org/10.1145/2601381.2601397","url":null,"abstract":"Automatic parallelization of models has been the \"Holy Grail\" of the PDES community for the last 20 years. In this paper we present LORAIN -- Low Overhead Runtime Assisted Instruction Negation -- a tool capable of automatic emission of a reverse event handler by the compiler. Upon detection of certain instructions, LORAIN is able to account for, and in many cases reverse, the computation without resorting to state-saving techniques. For our PDES framework, we use Rensselaer's Optimistic Simulation System (ROSS) coupled with the LLVM compiler to generate the reverse event handler. One of the primary contributions of this work is that LORAIN operates on the LLVM-generated Intermediate Representation (IR) as opposed to the model, high-level source code. Through information gleaned from the IR, LORAIN is able to analyze, instrument, and invert various operations and emit efficient reverse event handlers at the binary code level.\u0000 This preliminary work demonstrates the potential of this tool. We are able to reverse both the PHOLD model (a synthetic benchmark) as well as Fujimoto's airport model. Our results demonstrate that LORAIN-generated models are able to execute at a rate that is over 97% of hand-written, parallel model code performance.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127644422","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}
M&S-based analysis has been performed for simulation experiments of all possible input combinations as a 'what-if' analysis causing the simulation to be extremely time-consuming. To resolve this problem, this paper proposes a multi-fidelity M&S methodology for enhancing simulation speed while minimizing accuracy loss and maximizing model reusability, in the M&S-based analysis. Target systems of this methodology are continuous and discrete event system. The proposed multi-fidelity M&S methodology consists of 4 steps: 1) target model selection and Interest Region definition, 2) low-fidelity model development, 3) multi-fidelity model composition, 4) selected target model substitution. Also this methodology proposes structure of multi-fidelity model and its mathematical specifications for the third step. This methodology is applied without any modification of existing models and simulation engine for maximizing model reusability. Case study applies this methodology to Torpedo Tactics Simulation model and the Vehicle Allocation Simulation model. The result shows that simulation speed increases at least 1.21 times with 5% accuracy loss. We expect that this methodology will be applicable in various M&S-based analysis for enhancing simulation speed.
{"title":"Multi-fidelity modeling & simulation methodology for simulation speed up","authors":"S. Choi, Sun Ju Lee, T. Kim","doi":"10.1145/2601381.2601385","DOIUrl":"https://doi.org/10.1145/2601381.2601385","url":null,"abstract":"M&S-based analysis has been performed for simulation experiments of all possible input combinations as a 'what-if' analysis causing the simulation to be extremely time-consuming. To resolve this problem, this paper proposes a multi-fidelity M&S methodology for enhancing simulation speed while minimizing accuracy loss and maximizing model reusability, in the M&S-based analysis. Target systems of this methodology are continuous and discrete event system. The proposed multi-fidelity M&S methodology consists of 4 steps: 1) target model selection and Interest Region definition, 2) low-fidelity model development, 3) multi-fidelity model composition, 4) selected target model substitution. Also this methodology proposes structure of multi-fidelity model and its mathematical specifications for the third step. This methodology is applied without any modification of existing models and simulation engine for maximizing model reusability. Case study applies this methodology to Torpedo Tactics Simulation model and the Vehicle Allocation Simulation model. The result shows that simulation speed increases at least 1.21 times with 5% accuracy loss. We expect that this methodology will be applicable in various M&S-based analysis for enhancing simulation speed.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131038042","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}
Conducting simulation studies can mean to execute a multitude of parameter configurations, for each of these we may need to execute a vast number of replications, and each single replication may mean the need to process a significant amount of data. Here, we propose a stream-based architecture that aligns data processing and buffering with the actual data usage during simulation to make the most of available memory. This turns away from the first-write-then-read approach, often utilizing databases or plain files as temporary storage. Instead, data are processed on the fly. By introducing a processing graph, which distinguishes between buffering and processing nodes, a flexible analysis of simulation data is achieved. As the data are processed close to their generation, the developed architecture fits well to a distributed execution of simulation studies. We illustrate how the stream-based architecture integrates into simulation workflows.
{"title":"A stream-based architecture for the management and on-line analysis of unbounded amounts of simulation data","authors":"Johannes Schützel, Holger Meyer, A. Uhrmacher","doi":"10.1145/2601381.2601399","DOIUrl":"https://doi.org/10.1145/2601381.2601399","url":null,"abstract":"Conducting simulation studies can mean to execute a multitude of parameter configurations, for each of these we may need to execute a vast number of replications, and each single replication may mean the need to process a significant amount of data. Here, we propose a stream-based architecture that aligns data processing and buffering with the actual data usage during simulation to make the most of available memory. This turns away from the first-write-then-read approach, often utilizing databases or plain files as temporary storage. Instead, data are processed on the fly. By introducing a processing graph, which distinguishes between buffering and processing nodes, a flexible analysis of simulation data is achieved. As the data are processed close to their generation, the developed architecture fits well to a distributed execution of simulation studies. We illustrate how the stream-based architecture integrates into simulation workflows.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126054742","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}
Is computer science a science? This question has been asked since the inception of the field in the 1940s. Computer Science is certainly a science in the sense of a mature body of knowledge, including systematic method and practice. Many in the field have argued that the body of knowledge argument is sufficient. To provide a more substantive argument in favor of a science interpretation, other researchers have suggested that computer science is an artificial, experimental science not unlike economics. To build from, and complement prior views, we claim that computing is an empirical science, similar to that of physics or biology. Demonstrating this claim requires broadening the foundations of computing to include analog systems, and employing modeling and simulation as a fundamental approach toward observing computing in natural and artificial contexts. An example model of mixed discrete-event/continuous information is presented in support of this claim.
{"title":"Computing as model-based empirical science","authors":"P. Fishwick","doi":"10.1145/2601381.2601391","DOIUrl":"https://doi.org/10.1145/2601381.2601391","url":null,"abstract":"Is computer science a science? This question has been asked since the inception of the field in the 1940s. Computer Science is certainly a science in the sense of a mature body of knowledge, including systematic method and practice. Many in the field have argued that the body of knowledge argument is sufficient. To provide a more substantive argument in favor of a science interpretation, other researchers have suggested that computer science is an artificial, experimental science not unlike economics. To build from, and complement prior views, we claim that computing is an empirical science, similar to that of physics or biology. Demonstrating this claim requires broadening the foundations of computing to include analog systems, and employing modeling and simulation as a fundamental approach toward observing computing in natural and artificial contexts. An example model of mixed discrete-event/continuous information is presented in support of this claim.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124837949","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}
Simulation and emulation techniques are commonly used to evaluate the performance of complex networked systems. Simulation conveniently predicts the behavior of a complex networked system while usually requiring fewer simplifying model assumptions often necessary for theoretical analysis. In contrast, emulation does not need to re-implement the target real systems, so it may improve on the implementation efficiency of simulation while maintaining much of the realism of testbeds. A hybrid approach in which simulation nodes connect to emulation hosts can be used to combine the advantages of both approaches. In this paper, we propose integrating simulation with emulation using adaptive time dilation to evaluate system performance. If a simulator schedules its events in real time and the simulation time keeps up with the real time, then the hybrid system works very well and meets its deadlines. However, a heavily-loaded simulator can introduce significant simulation delays and thereby create situations where these delays impact the accuracy of the system. Our approach uses time dilation to reduce simulation delays and thus increasing the accuracy of the integrated simulation and emulation system. Our adaptive time dilation dynamically controls the time dilation factor to avoid system overloads for both the simulation and the emulation components and to improve the execution correctness of the hybrid system.
{"title":"Integrated simulation and emulation using adaptive time dilation","authors":"H. Lee, D. Thuente, M. Sichitiu","doi":"10.1145/2601381.2601384","DOIUrl":"https://doi.org/10.1145/2601381.2601384","url":null,"abstract":"Simulation and emulation techniques are commonly used to evaluate the performance of complex networked systems. Simulation conveniently predicts the behavior of a complex networked system while usually requiring fewer simplifying model assumptions often necessary for theoretical analysis. In contrast, emulation does not need to re-implement the target real systems, so it may improve on the implementation efficiency of simulation while maintaining much of the realism of testbeds. A hybrid approach in which simulation nodes connect to emulation hosts can be used to combine the advantages of both approaches. In this paper, we propose integrating simulation with emulation using adaptive time dilation to evaluate system performance. If a simulator schedules its events in real time and the simulation time keeps up with the real time, then the hybrid system works very well and meets its deadlines. However, a heavily-loaded simulator can introduce significant simulation delays and thereby create situations where these delays impact the accuracy of the system. Our approach uses time dilation to reduce simulation delays and thus increasing the accuracy of the integrated simulation and emulation system. Our adaptive time dilation dynamically controls the time dilation factor to avoid system overloads for both the simulation and the emulation components and to improve the execution correctness of the hybrid system.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"264 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133663542","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}
We present TimeKeeper: a simple lightweight approach to embedding Linux containers (LXC) in virtual time. Each container can be directed to progress in virtual time either more rapidly or more slowly than the physical wall clock time. As a result, interactions between an LXC and physical devices can be artificially scaled, e.g., to make a network appear to be ten times faster with respect to the software within the LXC than it actually is. Our approach also supports synchronized (in virtual time) emulation, by grouping LXCs together into an experiment where the virtual times of containers are kept synchronized, even when they advance at different speeds. This has direct application to the integration of emulation and simulation within a common framework.
{"title":"TimeKeeper: a lightweight virtual time system for linux","authors":"Jereme Lamps, D. Nicol, M. Caesar","doi":"10.1145/2601381.2601395","DOIUrl":"https://doi.org/10.1145/2601381.2601395","url":null,"abstract":"We present TimeKeeper: a simple lightweight approach to embedding Linux containers (LXC) in virtual time. Each container can be directed to progress in virtual time either more rapidly or more slowly than the physical wall clock time. As a result, interactions between an LXC and physical devices can be artificially scaled, e.g., to make a network appear to be ten times faster with respect to the software within the LXC than it actually is. Our approach also supports synchronized (in virtual time) emulation, by grouping LXCs together into an experiment where the virtual times of containers are kept synchronized, even when they advance at different speeds. This has direct application to the integration of emulation and simulation within a common framework.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133456845","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}
In this article we tackle transparent parallelization of Discrete Event Simulation (DES) models to be run on top of multi-core machines according to speculative schemes. The innovation in our proposal lies in that we consider a more general programming and execution model, compared to the one targeted by state of the art PDES platforms, where the boundaries of the state portion accessible while processing an event at a specific simulation object do not limit access to the actual object state, or to shared global variables. Rather, the simulation object is allowed to access (and alter) the state of any other object, thus causing what we term cross-state dependency. We note that this model exactly complies with typical (easy to manage) sequential-style DES programming, where a (dynamically-allocated) state portion of object A can be accessed by object B in either read or write mode (or both) by, e.g., passing a pointer to B as the payload of a scheduled simulation event. However, while read/write memory accesses performed in the sequential run are always guaranteed to observe (and to give rise to) a consistent snapshot of the state of the simulation model, consistency is not automatically guaranteed in case of parallelization and concurrent execution of simulation objects with cross-state dependencies. We cope with such a consistency issue, and its application-transparent support, in the context of parallel and optimistic executions. This is achieved by introducing an advanced memory management architecture, able to efficiently detect read/write accesses by concurrent objects to whichever object state in an application transparent manner, together with advanced synchronization mechanisms providing the advantage of exploiting parallelism in the underlying multi-core architecture while transparently handling both cross-state and traditional event-based dependencies. Our proposal targets Linux and has been integrated with the ROOT-Sim open source optimistic simulation platform, although its design principles, and most parts of the developed software, are of general relevance.
{"title":"Transparent multi-core speculative parallelization of DES models with event and cross-state dependencies","authors":"Alessandro Pellegrini, F. Quaglia","doi":"10.1145/2601381.2601398","DOIUrl":"https://doi.org/10.1145/2601381.2601398","url":null,"abstract":"In this article we tackle transparent parallelization of Discrete Event Simulation (DES) models to be run on top of multi-core machines according to speculative schemes. The innovation in our proposal lies in that we consider a more general programming and execution model, compared to the one targeted by state of the art PDES platforms, where the boundaries of the state portion accessible while processing an event at a specific simulation object do not limit access to the actual object state, or to shared global variables. Rather, the simulation object is allowed to access (and alter) the state of any other object, thus causing what we term cross-state dependency. We note that this model exactly complies with typical (easy to manage) sequential-style DES programming, where a (dynamically-allocated) state portion of object A can be accessed by object B in either read or write mode (or both) by, e.g., passing a pointer to B as the payload of a scheduled simulation event. However, while read/write memory accesses performed in the sequential run are always guaranteed to observe (and to give rise to) a consistent snapshot of the state of the simulation model, consistency is not automatically guaranteed in case of parallelization and concurrent execution of simulation objects with cross-state dependencies. We cope with such a consistency issue, and its application-transparent support, in the context of parallel and optimistic executions. This is achieved by introducing an advanced memory management architecture, able to efficiently detect read/write accesses by concurrent objects to whichever object state in an application transparent manner, together with advanced synchronization mechanisms providing the advantage of exploiting parallelism in the underlying multi-core architecture while transparently handling both cross-state and traditional event-based dependencies. Our proposal targets Linux and has been integrated with the ROOT-Sim open source optimistic simulation platform, although its design principles, and most parts of the developed software, are of general relevance.","PeriodicalId":255272,"journal":{"name":"SIGSIM Principles of Advanced Discrete Simulation","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-05-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114494105","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：