{"title":"Operating system impact on trace-driven simulation","authors":"J. Casmira, John Fraser, D. Kaeli, W. Meleis","doi":"10.1109/SIMSYM.1998.668444","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668444","url":null,"abstract":"Trace-driven simulation is commonly used by the computer architecture research community to pursue answers to a wide variety of architectural design issues. Traces taken from benchmark execution have been extensively studied to optimize the design of pipelines, branch predictors, and especially cache memories. Today's computer designs have been optimized based on the characteristics of these benchmarks. One important aspect that has been ignored in a majority of these trace-driven studies is the effect of the operating system interacting with the benchmark. It has been acknowledged that operating system overhead can introduce a level of interference that can limit the benefits of new designs. The reason why the operating system has been, for the most part, ignored in these studies is the lack of readily available tools that can generate kernel-laden traces. In this paper we describe two tracing systems that allow the capture of operating system and application traces. We have captured traces of both benchmark and commercial applications using these tools. We show the effects of including operating system activity while studying various architectural design tradeoffs. We have found that the operating system can increase the number of instructions captured in the trace (for a given time sample) by as much as 100X, can significantly modify the instruction mix, and produce different ranges of reference locality. The results of this study show that the inherent characteristics of the applications will dictate the amount of overhead introduced by the operating system.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116897214","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":"Network traffic characterization of distributed database applications","authors":"Chatchai Khunboa, S. Banerjee","doi":"10.1109/SIMSYM.1998.668455","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668455","url":null,"abstract":"The authors' efforts in characterizing the traffic of distributed database systems are described. This is an important class of applications considering that database or transaction processing systems form the core of several key commercial applications. Specific traffic source models for this application class have not been developed which is imperative for their successful deployment in emerging networks. The main contribution of the paper is to present the relationship between a distributed database application scenario and the corresponding network traffic characteristics, thus providing guidelines to these applications to compute their traffic descriptors accurately.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116933611","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":"Applying intelligent agent technology as the platform for simulation","authors":"Gary S. H. Tan, Kwok-Leong Hui","doi":"10.1109/SIMSYM.1998.668486","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668486","url":null,"abstract":"Simulation is one of the most powerful analysis tools available for the design and operation of complex processes or systems. In an increasingly competitive world, simulation has become a very powerful tool for the planning, design and control of systems. In recent years, there has been a strong revival and advancement in the area of intelligent agent technology in artificial intelligence (AI) and the authors foresee a strong application of intelligent agent technology in simulation. Hence, the paper attempts to put forward an architecture for performing simulation using intelligent agent technology.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114499967","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":"Performance impacts of superscalar microarchitecture on SOM execution","authors":"O. Hammami","doi":"10.1109/SIMSYM.1998.668489","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668489","url":null,"abstract":"Neural network simulations are notorious for being very time and resource consuming. However, although general purpose microprocessors have improved the performance of these simulations, little is known about which microarchitecture features contribute the most to this performance improvement. In this context, the paper analyzes the performance impact of various microarchitectural mechanisms found in current superscalar microprocessors on the execution of a famous neural network, the SOM algorithm. The conclusion is that the SOM algorithm does not fully benefit from the sophisticated hardware support existing in a state of the art superscalar machine. It is especially true of the memory hierarchy as well as the branch prediction mechanisms.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129476282","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 knowledge-based simulation","authors":"Elena B. Durán, R. Costaguta","doi":"10.1109/SIMSYM.1998.668485","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668485","url":null,"abstract":"The convergence of the representation based methods used in artificial intelligence and simulation has originated the emergence of knowledge-based simulation. The objective of this new approach is to achieve more powerful and understandable simulations, which will allow the simulation developers to construct, validate, develop and maintain simulations that can model a wider range of complex systems, and which will allow the users to interact with the simulation in order to achieve a greater comprehension of the modeling phenomenon. There is a definition of the possible contributions that the expert systems can give to the different life cycle stages of the discrete event simulation. Specifically, the authors propose an example of a discrete event system, in particular a waiting phenomenon, and consequently constructing production rules to be applied in model development, parameter estimation and results analysis and interpretation.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"200 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116153220","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":"Effect of local retransmission at wireless access points on the round trip time estimation of TCP","authors":"K. Ratnam, I. Matta","doi":"10.1109/SIMSYM.1998.668481","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668481","url":null,"abstract":"The transmission control protocol (TCP) has been mainly designed assuming a relatively reliable wireline network. It is known to perform poorly in the presence of wireless links because of its basic assumption that any loss of a data segment is due to congestion and consequently it invokes congestion control measures. However, on wireless access links, a large number of segment losses will occur more often because of wireless link errors or host mobility. For this reason, many proposals have recently appeared to improve TCP performance in such environment. They usually rely on the wireless access points (base stations) to locally re-transmit the data in order to hide wireless losses from TCP. The authors use extensive simulations to evaluate TCP performance in the presence of congestion and wireless losses when the base station employs one of two proposals, namely Snoop and WTCP. The results show that WTCP significantly improves the throughput of TCP connections due to its unique feature of hiding the time spent by the base station to locally recover from wireless link errors so that TCP's round trip time estimation at the source is not affected. This proved to be critical since otherwise the ability of the source to effectively detect congestion in the fixed wireline network is hindered.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"42 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120897765","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":"Fair arbitration in point-to-point networks","authors":"D. Avresky, V. Shurbanov, R. Horst","doi":"10.1109/SIMSYM.1998.668433","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668433","url":null,"abstract":"In this paper we introduce and evaluate a method for optimizing the router arbitration policy in point-to-point networks. These networks use multiple high speed, low-cost routers to rapidly switch data directly between data sources and destinations. We assume that routers use an arbitration algorithm that has the ability to assign different priorities to different links thus controlling the distribution of available bandwidth among channels. We propose an analytical method that allows us to determine the correct priorities for channels so as to achieve fair arbitration thus minimizing latency, increasing throughput and eliminating tree saturation. In our study we used data generated by a simulation tool to validate the proposed analytical model for determining the weights (priorities) of router ports and to evaluate the resulting performance improvements. Link usage statistics were recorded by the simulator for the purpose of performing a detailed investigation of congestion effects and especially tree saturation. The method was applied to the ServerNet/sup TM/ System Area Network (SAN), developed by Tandem Computers Inc.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127467581","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":"An adaptive synchronization protocol for parallel discrete event simulation","authors":"K. Bisset","doi":"10.1109/SIMSYM.1998.668421","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668421","url":null,"abstract":"Simulation, especially discrete event simulation (DES), is used in a variety of disciplines where numerical methods are difficult or impossible to apply. One problem with this method is that a sufficiently detailed simulation may take hours or days to execute, and multiple runs may be needed in order to generate the desired results. Parallel discrete event simulation (PDES) has been explored for many years as a method to decrease the time taken to execute a simulation. Many protocols have been developed which work well for particular types of simulations, but perform poorly when used for other types of simulations. Often it is difficult to know a priori whether a particular protocol is appropriate for a given problem. In this work, an adaptive synchronization method (ASM) is developed which works well on an entire spectrum of problems. The ASM determines, using artificial neural network (ANN), the likelihood that a particular event is safe to process.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121843376","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":"A tool for statistical analysis of hard real-time scheduling algorithms","authors":"J. Goossens, Christian Hernalsteen","doi":"10.1109/SIMSYM.1998.668435","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668435","url":null,"abstract":"In this paper we shall consider a simulation language (and a tool) for the simulation of scheduling algorithms for real-time systems. In the first part of this paper we recall some notion on real-time systems and their scheduling algorithms. The second part of the work presents our tool. We first present shortly the simulation kernel which is well suited to simulate the behavior of a single real-time system with some scheduling algorithm. Next, we show the limitations of the simulation kernel and the interest to simulate scheduling algorithms on a large number of task sets in order to be able to perform statistical analysis on scheduling algorithms. We then present the extensions of our tool and the mechanisms we have added for statistical analysis.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127227609","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":"Efficient partitioning method for distributed logic simulation of VLSI circuits","authors":"A. Guettaf, P. Bazargan-Sabet","doi":"10.1109/SIMSYM.1998.668488","DOIUrl":"https://doi.org/10.1109/SIMSYM.1998.668488","url":null,"abstract":"Distributed simulation is expected to provide a significant speed up of simulation run time. Partitioning and load balancing are very influential factors for speed up. The paper presents an efficient partitioning method for distributed VLSI circuits simulation. The main features of this method are the use of a logic replication algorithm. A realistic cost function based on precalculated activity of the circuit using a probabilistic algorithm, and a the balance between execution cost and communication cost. A distributed simulator based on a conservative synchronization method has been used to evaluate the performance of the partitioning.","PeriodicalId":339060,"journal":{"name":"Proceedings 31st Annual Simulation Symposium","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1998-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128051199","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}
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
