A warm welcome to SIGSIM PADS'17, the 5th ACM SIGSIM Conference on Principles of Advanced Discrete Simulation. SIGSIM PADS is the flagship conference of ACM's Special Interest Group on Simulation and Modeling (SIGSIM). It provides a unique forum for reporting and discussing research results and important topics of interest to the modeling and simulation (M&S) community. The annual PADS conference has a long history dating back to 1985. Over the years PADS has broadened its scope beyond its origins in parallel and distributed simulation and now encompasses virtually all research that lies at the intersection of the computer science and the M&S fields. Built on its strong history, PADS became the ACM SIGSIM's flagship conference and renamed as SIGSIM PADS in 2013. This year is the 5th edition of the conference under its new brand name. PADS was first held in Singapore in 2006. We are very pleased that after 11 years PADS returns to Singapore under its new brand name. Singapore is a dynamic city rich in contrast and color, where you will find a harmonious blend of culture, cuisine, arts and architecture. With its friendly and welcome people, state-of-the-art infrastructure and spectacular events, Singapore has everything to make your stay a most memorable experience. This year's SIGSIM PADS received a large number of submissions, further strengthening its status as a leading conference in its area. All papers submitted to the conference were rigorously reviewed with at least 3 reviewers for each paper. We would like to thank the program committee and additional reviewers for their diligent efforts to provide timely, critical reviews and feedback to the authors. The two and half day conference consists of 2 keynote speeches, 2 invited talks, and presentations of 21 full papers and 2 short papers. The first keynote will be given by Dr. Marc-Oliver Gewaltig on "Towards Simulating the Human Brian", and the second keynote will be given by Prof. Young-Jun Son on "An Integrated Human Decision Making Model under Extended Belief-Desire-Intention Framework". We hope that you enjoy this exciting program that we have arranged for you. There are 8 students participating in this year's Ph.D. Colloquium. They will give brief presentations as well as showing posters concerning their research. Prof. Philip Wilsey will give a keynote presentation for the PhD Colloquium. In keeping with the PADS tradition, a Best Paper Committee will select the SIGSIM PADS'17 Best Paper Award from the most highly ranked papers by reviewers. The candidates for this year's best paper are (in no particular order): Julius Higiro, Meseret Gebre and Dhananjai Rao. "Multi-tier Priority Queues & 2-tier Ladder Queue for Managing Pending Events in Sequential & Optimistic Parallel Simulations" Yulin Wu, Xiangting Hou, Wenjun Tan, Zengxiang Li and Wentong Cai. "Efficient Parallel Simulation over Social Contact Network with Skewed Degree Distribution" Md Shafiur Rahman, Nael Abu-Gha
热烈欢迎参加SIGSIM PADS'17，第五届ACM SIGSIM高级离散仿真原理会议。SIGSIM PADS是ACM仿真与建模特别兴趣小组(SIGSIM)的旗舰会议。它为报告和讨论研究结果以及建模和仿真(M&S)社区感兴趣的重要主题提供了一个独特的论坛。每年一度的pad会议历史悠久，可以追溯到1985年。多年来，PADS已经扩展了其范围，超越了其起源的并行和分布式仿真，现在几乎涵盖了计算机科学和M&S领域交叉的所有研究。基于其悠久的历史，PADS成为ACM SIGSIM的旗舰会议，并于2013年更名为SIGSIM PADS。今年是该会议的第五届会议，该会议以其新品牌命名。PADS首届会议于2006年在新加坡举行。我们很高兴在时隔11年后，PADS以新的品牌名称重返新加坡。新加坡是一个充满活力和色彩的城市，在这里你会发现文化、美食、艺术和建筑的和谐融合。热情好客的新加坡人民、最先进的基础设施和壮观的活动，新加坡的一切都将使您的住宿成为最难忘的经历。今年的SIGSIM PADS收到了大量的提交，进一步加强了其作为该领域领先会议的地位。所有提交给会议的论文都经过严格的审查，每篇论文至少有3名审稿人。我们要感谢项目委员会和其他审稿人的辛勤努力，为作者提供及时、关键的审查和反馈。会议为期两天半，包括2场主题演讲，2场特邀演讲，21篇论文和2篇短文。第一个主题演讲将由Marc-Oliver Gewaltig博士作“模拟人类大脑”，第二个主题演讲将由Young-Jun Son教授作“扩展信念-欲望-意图框架下的综合人类决策模型”。我们希望你喜欢我们为你安排的这个激动人心的节目。有8名学生参加了今年的博士讨论会。他们将作简短的报告，并展示有关他们研究的海报。Philip Wilsey教授将在博士研讨会上作主题演讲。为了与PADS的传统保持一致，最佳论文委员会将从评审者排名最高的论文中选出SIGSIM PADS的17个最佳论文奖。今年最佳论文的候选人是(排名不分先后):Julius Higiro, Meseret Gebre和Dhananjai Rao。吴玉林，侯祥婷，谭文军，李增祥，蔡文通，“时序与乐观并行仿真中并发事件管理的多层优先级队列和二层阶梯队列”。Md Shafiur Rahman, Nael Abu-Ghazaleh和Walid Najjar，“具有偏斜度分布的社会联系网络的有效并行模拟”。“PDES-A: fpga的并行离散事件仿真加速器”最佳论文奖将在会议的宴会上公布。我们代表SIGSIM PADS'16项目主席，很高兴地宣布，SIGSIM PADS'16最佳论文的获奖者是Mirko Stoffers, Daniel Schemmel, Oscar Soria Dustmann和Klaus Wehrle的“在非纯语言中实现参数研究的自动记忆”。
{"title":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","authors":"Wentong Cai, Teo Yong Meng, P. Wilsey, Kevin Jin","doi":"10.1145/3064911","DOIUrl":"https://doi.org/10.1145/3064911","url":null,"abstract":"A warm welcome to SIGSIM PADS'17, the 5th ACM SIGSIM Conference on Principles of Advanced Discrete Simulation. SIGSIM PADS is the flagship conference of ACM's Special Interest Group on Simulation and Modeling (SIGSIM). It provides a unique forum for reporting and discussing research results and important topics of interest to the modeling and simulation (M&S) community. The annual PADS conference has a long history dating back to 1985. Over the years PADS has broadened its scope beyond its origins in parallel and distributed simulation and now encompasses virtually all research that lies at the intersection of the computer science and the M&S fields. Built on its strong history, PADS became the ACM SIGSIM's flagship conference and renamed as SIGSIM PADS in 2013. This year is the 5th edition of the conference under its new brand name. \u0000 \u0000PADS was first held in Singapore in 2006. We are very pleased that after 11 years PADS returns to Singapore under its new brand name. Singapore is a dynamic city rich in contrast and color, where you will find a harmonious blend of culture, cuisine, arts and architecture. With its friendly and welcome people, state-of-the-art infrastructure and spectacular events, Singapore has everything to make your stay a most memorable experience. \u0000 \u0000This year's SIGSIM PADS received a large number of submissions, further strengthening its status as a leading conference in its area. All papers submitted to the conference were rigorously reviewed with at least 3 reviewers for each paper. We would like to thank the program committee and additional reviewers for their diligent efforts to provide timely, critical reviews and feedback to the authors. The two and half day conference consists of 2 keynote speeches, 2 invited talks, and presentations of 21 full papers and 2 short papers. The first keynote will be given by Dr. Marc-Oliver Gewaltig on \"Towards Simulating the Human Brian\", and the second keynote will be given by Prof. Young-Jun Son on \"An Integrated Human Decision Making Model under Extended Belief-Desire-Intention Framework\". We hope that you enjoy this exciting program that we have arranged for you. \u0000 \u0000There are 8 students participating in this year's Ph.D. Colloquium. They will give brief presentations as well as showing posters concerning their research. Prof. Philip Wilsey will give a keynote presentation for the PhD Colloquium. \u0000 \u0000In keeping with the PADS tradition, a Best Paper Committee will select the SIGSIM PADS'17 Best Paper Award from the most highly ranked papers by reviewers. The candidates for this year's best paper are (in no particular order): \u0000Julius Higiro, Meseret Gebre and Dhananjai Rao. \"Multi-tier Priority Queues & 2-tier Ladder Queue for Managing Pending Events in Sequential & Optimistic Parallel Simulations\" \u0000Yulin Wu, Xiangting Hou, Wenjun Tan, Zengxiang Li and Wentong Cai. \"Efficient Parallel Simulation over Social Contact Network with Skewed Degree Distribution\" \u0000Md Shafiur Rahman, Nael Abu-Gha","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129805087","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}
The choice of data structure for managing and processing pending events in timestamp priority order plays a critical role in achieving good performance of sequential and parallel Discrete Event Simulation (DES). Accordingly, we propose and evaluate the effectiveness of our novel multi-tiered (2 and 3 tier) data structures and our 2-tier Ladder Queue, for both sequential and optimistic parallel simulations, on distributed memory platforms. Our assessments use (a fine-tuned version of) the Ladder Queue, which has shown to outperform many other data structures for DES. The experimental results based on 2,500 configurations of PHOLD benchmark show that our 3-tier heap and 2-tier ladder queue outperform the Ladder Queue by 10% to 50% in simulations, particularly those with higher concurrency per Logical Process (LP), in both sequential and Time Warp synchronized parallel simulations.
{"title":"Multi-tier Priority Queues and 2-tier Ladder Queue for Managing Pending Events in Sequential and Optimistic Parallel Simulations","authors":"Julius Higiro, Meseret Gebre, D. Rao","doi":"10.1145/3064911.3064921","DOIUrl":"https://doi.org/10.1145/3064911.3064921","url":null,"abstract":"The choice of data structure for managing and processing pending events in timestamp priority order plays a critical role in achieving good performance of sequential and parallel Discrete Event Simulation (DES). Accordingly, we propose and evaluate the effectiveness of our novel multi-tiered (2 and 3 tier) data structures and our 2-tier Ladder Queue, for both sequential and optimistic parallel simulations, on distributed memory platforms. Our assessments use (a fine-tuned version of) the Ladder Queue, which has shown to outperform many other data structures for DES. The experimental results based on 2,500 configurations of PHOLD benchmark show that our 3-tier heap and 2-tier ladder queue outperform the Ladder Queue by 10% to 50% in simulations, particularly those with higher concurrency per Logical Process (LP), in both sequential and Time Warp synchronized parallel simulations.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125428241","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":"Session details: Paper Session 2 Parallel Simulation II","authors":"D. Rao","doi":"10.1145/3254051","DOIUrl":"https://doi.org/10.1145/3254051","url":null,"abstract":"","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133987925","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":"Session details: Paper Session 5 Simulation Application I: Networking and Communication","authors":"N. Abu-Ghazaleh","doi":"10.1145/3254054","DOIUrl":"https://doi.org/10.1145/3254054","url":null,"abstract":"","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131241469","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}
Experience and intuition indicate that both synchronization and the mapping of workload to processors has significant impact on overall performance. However, the behavior of parallel simulations is quite complex, and the inter-relationships between workload mapping and the synchronization overheads need mathematical explanation. This paper develops a performance model of a parallel simulation that is synchronized using composite synchronization. We use this model to help explain how mapping decisions and a synchronization tuning parameter impacts synchronization overhead, and hence performance. The observations we make should inform designers of algorithms to map conservatively synchronized parallel simulations to the available computing platform.
{"title":"A Performance Model of Composite Synchronization","authors":"D. Nicol","doi":"10.1145/3064911.3069396","DOIUrl":"https://doi.org/10.1145/3064911.3069396","url":null,"abstract":"Experience and intuition indicate that both synchronization and the mapping of workload to processors has significant impact on overall performance. However, the behavior of parallel simulations is quite complex, and the inter-relationships between workload mapping and the synchronization overheads need mathematical explanation. This paper develops a performance model of a parallel simulation that is synchronized using composite synchronization. We use this model to help explain how mapping decisions and a synchronization tuning parameter impacts synchronization overhead, and hence performance. The observations we make should inform designers of algorithms to map conservatively synchronized parallel simulations to the available computing platform.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121958685","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 the particle-based simulation of cell-biological systems in continuous space, a key performance bottleneck is the computation of all possible intersections between particles. These typically rely for collision detection on solid sphere approaches. The behavior of cell biological systems is influenced by dynamic hierarchical nesting, such as the forming of, the transport within, and the merging of vesicles. Existing collision detection algorithms are found not to be designed for these types of spatial cell-biological models, because nearly all existing high performance parallel algorithms are focusing on solid sphere interactions. The known algorithms for solid sphere intersections return more intersections than actually occur with nested hollow spheres. Here we define a new problem of computing the intersections among arbitrarily nested hollow spheres of possibly different sizes, thicknesses, positions, and nesting levels. We describe a new algorithm designed to solve this nested hollow sphere intersection problem and implement it for parallel execution on graphical processing units (GPUs). We present first results about the runtime performance and scaling to hundreds of thousands of spheres, and compare the performance with that from a leading solid object intersection package also running on GPUs.
{"title":"Efficient Simulation of Nested Hollow Sphere Intersections: for Dynamically Nested Compartmental Models in Cell Biology","authors":"Till Köster, K. Perumalla, A. Uhrmacher","doi":"10.1145/3064911.3064920","DOIUrl":"https://doi.org/10.1145/3064911.3064920","url":null,"abstract":"In the particle-based simulation of cell-biological systems in continuous space, a key performance bottleneck is the computation of all possible intersections between particles. These typically rely for collision detection on solid sphere approaches. The behavior of cell biological systems is influenced by dynamic hierarchical nesting, such as the forming of, the transport within, and the merging of vesicles. Existing collision detection algorithms are found not to be designed for these types of spatial cell-biological models, because nearly all existing high performance parallel algorithms are focusing on solid sphere interactions. The known algorithms for solid sphere intersections return more intersections than actually occur with nested hollow spheres. Here we define a new problem of computing the intersections among arbitrarily nested hollow spheres of possibly different sizes, thicknesses, positions, and nesting levels. We describe a new algorithm designed to solve this nested hollow sphere intersection problem and implement it for parallel execution on graphical processing units (GPUs). We present first results about the runtime performance and scaling to hundreds of thousands of spheres, and compare the performance with that from a leading solid object intersection package also running on GPUs.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"236 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114396808","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}
When designing, implementing and executing large-scale distributed simulation codes it is well-known the so-called optimistic time synchronization methods often lead to better overall performance as compared to the so-called conservative methods. Particularly when the model being simulated suffers from small lookahead between two or more of the logical processes working on the overall simulation execution. However, the design and implementation of simulation model codes intended to be executed using optimistic approach is often daunting, especially when working with existing codes which were designed and implemented to be used in a conservative approach. When executing simulation events, optimistic simulation models must first save any existing state in the model prior to executing codes that change that state, and furthermore must implement rollback code which restores that state to the original value. The necessity for such extra software functionality is often a barrier to the complete implementation of an optimistic-based simulation model and the corresponding improved performance. However, recent work by Lawrence Livermore National Laboratory (LLNL) shows promise toward easing the burden of state-saving and rollback which will ultimately lead to simulation models that include state-saving and rollback automatically with little or no extra effort on the part of the model developer. This work, known as Backstroke, uses the C or C++ source code of the simulation model and generates both the forward-direction state-saving and the reverse-direction rollback prior to executing any simulation event. In this work, we demonstrate that the Network Simulator 3 (ns-3) model code, instrumented by Backstroke, can successfully execute scenario to some arbitrary time T, and then roll back to simulation time T0, restoring previously saved state for all modified ns-3 events and resulting in identical model state to that created in the original prior to the execution of the very first model event. Finally, we measure and report on the amount of wall-clock time overhead incurred by the ns-3 execution required by the Backstroke state saving and restoration, and compare that time to the execution time of the original unmodified ns-3 model codes.
{"title":"Automatic State Saving and Rollback in ns-3","authors":"Euna Kim, G. Riley","doi":"10.1145/3064911.3064925","DOIUrl":"https://doi.org/10.1145/3064911.3064925","url":null,"abstract":"When designing, implementing and executing large-scale distributed simulation codes it is well-known the so-called optimistic time synchronization methods often lead to better overall performance as compared to the so-called conservative methods. Particularly when the model being simulated suffers from small lookahead between two or more of the logical processes working on the overall simulation execution. However, the design and implementation of simulation model codes intended to be executed using optimistic approach is often daunting, especially when working with existing codes which were designed and implemented to be used in a conservative approach. When executing simulation events, optimistic simulation models must first save any existing state in the model prior to executing codes that change that state, and furthermore must implement rollback code which restores that state to the original value. The necessity for such extra software functionality is often a barrier to the complete implementation of an optimistic-based simulation model and the corresponding improved performance. However, recent work by Lawrence Livermore National Laboratory (LLNL) shows promise toward easing the burden of state-saving and rollback which will ultimately lead to simulation models that include state-saving and rollback automatically with little or no extra effort on the part of the model developer. This work, known as Backstroke, uses the C or C++ source code of the simulation model and generates both the forward-direction state-saving and the reverse-direction rollback prior to executing any simulation event. In this work, we demonstrate that the Network Simulator 3 (ns-3) model code, instrumented by Backstroke, can successfully execute scenario to some arbitrary time T, and then roll back to simulation time T0, restoring previously saved state for all modified ns-3 events and resulting in identical model state to that created in the original prior to the execution of the very first model event. Finally, we measure and report on the amount of wall-clock time overhead incurred by the ns-3 execution required by the Backstroke state saving and restoration, and compare that time to the execution time of the original unmodified ns-3 model codes.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126925075","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":"Session details: Paper Session 4 GPU and Hardware Acceleration","authors":"F. Quaglia","doi":"10.1145/3254053","DOIUrl":"https://doi.org/10.1145/3254053","url":null,"abstract":"","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"115 19","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113935137","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}
Understanding the human brain is still one of the biggest scientific challenges. The European Human Brain Project tries to tackle this challenge, by integrating a wide range of neuroscientific data into large multi-scale models and simulations of the brain. In this talk, I will highlight recent results and challenges that we face in our endeavour to reconstruct and simulate models of entire brains. The human brain is comprised of 80 billion neurons and 100*1012 synapses, each with dynamic properties that are governed by many differential equations. Representing the dynamic state of a complete human brain thus is still outside the reach of even the largest super-computers. Models of a mouse brain, still comprise 75 million neurons and 80 billion connections, but these are accessible with model supercomputers. In the first part of the talk, I will outline how high-resolution imaging data can be used to semi-automatically reconstruct 3D the positions of different neuron types as well as their connections. Next, I will discuss the challenges of representing and simulating such large-scale models using hybrid time and event driven simulation techniques. Finally, I will discuss applications of large-scale brain models for neuroscience, medicine, robotics and computing technology.
{"title":"Towards Simulating the Human Brain","authors":"M. Gewaltig","doi":"10.1145/3064911.3064935","DOIUrl":"https://doi.org/10.1145/3064911.3064935","url":null,"abstract":"Understanding the human brain is still one of the biggest scientific challenges. The European Human Brain Project tries to tackle this challenge, by integrating a wide range of neuroscientific data into large multi-scale models and simulations of the brain. In this talk, I will highlight recent results and challenges that we face in our endeavour to reconstruct and simulate models of entire brains. The human brain is comprised of 80 billion neurons and 100*1012 synapses, each with dynamic properties that are governed by many differential equations. Representing the dynamic state of a complete human brain thus is still outside the reach of even the largest super-computers. Models of a mouse brain, still comprise 75 million neurons and 80 billion connections, but these are accessible with model supercomputers. In the first part of the talk, I will outline how high-resolution imaging data can be used to semi-automatically reconstruct 3D the positions of different neuron types as well as their connections. Next, I will discuss the challenges of representing and simulating such large-scale models using hybrid time and event driven simulation techniques. Finally, I will discuss applications of large-scale brain models for neuroscience, medicine, robotics and computing technology.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127105480","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}
Romolo Marotta, Mauro Ianni, Alessandro Pellegrini, F. Quaglia
Emerging share-everything Parallel Discrete Event Simulation (PDES) platforms rely on worker threads fully sharing the workload of events to be processed. These platforms require efficient event pool data structures enabling high concurrency of extraction/insertion operations. Non-blocking event pool algorithms are raising as promising solutions for this problem. However, the classical non-blocking paradigm leads concurrent conflicting operations, acting on a same portion of the event pool data structure, to abort and then retry. In this article we present a conflict-resilient non-blocking calendar queue that enables conflicting dequeue operations, concurrently attempting to extract the minimum element, to survive, thus improving the level of scalability of accesses to the hot portion of the data structure---namely the bucket to which the current locality of the events to be processed is bound. We have integrated our solution within an open source share-everything PDES platform and report the results of an experimental analysis of the proposed concurrent data structure compared to some literature solutions.
{"title":"A Conflict-Resilient Lock-Free Calendar Queue for Scalable Share-Everything PDES Platforms","authors":"Romolo Marotta, Mauro Ianni, Alessandro Pellegrini, F. Quaglia","doi":"10.1145/3064911.3064926","DOIUrl":"https://doi.org/10.1145/3064911.3064926","url":null,"abstract":"Emerging share-everything Parallel Discrete Event Simulation (PDES) platforms rely on worker threads fully sharing the workload of events to be processed. These platforms require efficient event pool data structures enabling high concurrency of extraction/insertion operations. Non-blocking event pool algorithms are raising as promising solutions for this problem. However, the classical non-blocking paradigm leads concurrent conflicting operations, acting on a same portion of the event pool data structure, to abort and then retry. In this article we present a conflict-resilient non-blocking calendar queue that enables conflicting dequeue operations, concurrently attempting to extract the minimum element, to survive, thus improving the level of scalability of accesses to the hot portion of the data structure---namely the bucket to which the current locality of the events to be processed is bound. We have integrated our solution within an open source share-everything PDES platform and report the results of an experimental analysis of the proposed concurrent data structure compared to some literature solutions.","PeriodicalId":341026,"journal":{"name":"Proceedings of the 2017 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123179681","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：