Search

Computer architecture simulation tools are essential for implementing and evaluating new ideas in the domain and can be useful for understanding the behavior of programs and finding microarchitectural bottlenecks. One particularly important part of almost any processor is the cache hierarchy. While some simulators support simulating a whole processor, including the cache hierarchy, cores, and on-chip interconnect, others may only support simulating the cache hierarchy. This survey provides a detailed discussion on 28 CPU cache simulators, including popular or recent simulators. We compare between all of these simulators in four different ways: major design characteristics, support for specific cache design features, support for specific cache-related metrics, and validation methods and efforts. The strengths and shortcomings of each simulator and major issues that are common to all simulators are highlighted. The information presented in this survey was collected from many different sources, including research papers, documentations, source code bases, and others. This survey is potentially useful for both users and developers of cache simulators. To the best of our knowledge, this is the first comprehensive survey on cache simulation tools. [ABSTRACT FROM AUTHOR]

Networking research often relies on simulation in order to test and evaluate new ideas. An important requirement of this process is that results must be reproducible so that other researchers can replicate, validate, and extend existing work. We look at the landscape of simulators for research in peer-to-peer (P2P) networks by conducting a survey of a combined total of over 280 papers from before and after 2007 (the year of the last survey in this area), and comment on the large quantity of research using bespoke, closedsource simulators. We propose a set of criteria that P2P simulators should meet, and poll the P2P research community for their agreement. We aim to drive the community towards performing their experiments on simulators that allow for others to validate their results [ABSTRACT FROM AUTHOR]

The goal is to survey dependability modeling and analysis of software and systems specified with UML, with focus on reliability, availability, maintainability, and safety (RAMS). From the literature published in the last decade, 33 approaches presented in 43 papers were identified. They are evaluated according to three sets of criteria regarding UML modeling issues, addressed dependability characteristics, and quality assessment of the surveyed approaches. The survey shows that more works are devoted to reliability and safety, fewer to availability and maintainability, and none to integrity. Many methods support early life-cycle phases (from requirements to design). More research is needed for tool development to automate the derivation of analysis models and to give feedback to designers. [ABSTRACT FROM AUTHOR]

Benchmarking is how the performance of a computing system is determined. Surprisingly, even for classical computers this is not a straightforward process. One must choose the appropriate benchmark and metrics to extractmeaningful results. Different benchmarks test the system in different ways, and each individualmetric may or may not be of interest. Choosing the appropriate approach is tricky. The situation is even more open ended for quantum computers, where there is a wider range of hardware, fewer established guidelines, and additional complicating factors. Notably, quantum noise significantly impacts performance and is difficult to model accurately. Here, we discuss benchmarking of quantum computers from a computer architecture perspective and provide numerical simulations highlighting challenges that suggest caution. [ABSTRACT FROM AUTHOR]

Since the early 1970s, the application of queueing network models to computer system performance analysis has generated considerable interest. The period since has seen considerable progress in advances both in the fundamental theory and in practical experience. Other papers in this special issue have surveyed the state of the art. This note attempts to provide a condensed statement as to where we are and, as much as possible, to indicate areas of future research with high potential. [ABSTRACT FROM AUTHOR]

The two primary issues in choosing a computing system model are credibility of the model and cost of developing and solving the model. Credibility is determined by 1) the experience and biases of the persons using the model, 2) the extent to which the model represents system features, and 3) the accuracy of the solution technique. Queueing network models are widely used because they have proven effective and are inexpensive to solve. However, most queueing network models make strong assumptions to assure an exact numerical solution. When such assumptions severely affect credibility, simulation or other approaches are used, in spite of their relatively high cost. It is the contention of this paper that queueing network models with credible assumptions can be solved approximately to provide credible performance estimates at low cost. This contention is supported by examples of approximate solutions of queueing network models. Two major approaches to approximate solution, aggregation (decomposition) and diffusion, are discussed. [ABSTRACT FROM AUTHOR]

Encouraging experiences gained from the Michigan Terminal System (MTS) are reported as a response to a recent survey of multiprocessing organizations by Enslow [7]. This paper discusses the importance of the degree of symmetry within multiport system organizations, the ease of implementation experienced with the symmetrically organized operating system nucleus for MTS, throughput statistics for MTS which are better than those generally documented in the literature, and underlying reasons that might account for the impressive statistics. Since some of these results are at variance with the conventional wisdom concerning multiprocessor systems, a strong recommendation is made for additional modeling and measurement analysis based upon these experiences. [ABSTRACT FROM AUTHOR]

Both simulation and analytic models of computer systems can be very useful for predicting the performance of proposed new systems or proposed changes to existing systems. Unfortunately, many potential users of models are reluctant to use them because of the complexity of many such models and the difficulty of relating the model to the real system. This tutorial paper leads the reader through the development and use of an easily understood analytic model. This is then placed in context with a class of similar analytic models. In spite of the simplicity of these models they have proved useful and quite accurate in predicting performance (utilization, throughput, and response) using only the most basic system data as input. These parameters can either be estimates or measurements from a running system. The model equations and assumptions are defined, and a detailed case study is presented as an example of their use. [ABSTRACT FROM AUTHOR]

Modeling and predicting player behavior is of the utmost importance in developing games. Experience has proven that, while theory-driven approaches are able to comprehend and justify a model’s choices, such models frequently fail to encompass necessary features because of a lack of insight of the model builders. In contrast, data-driven approaches rely much less on expertise, and thus offer certain potential advantages. Hence, this study conducts a systematic review of the extant research on data-driven approaches to game player modeling. To this end, we have assessed experimental studies of such approaches over a nine-year period, from 2008 to 2016; this survey yielded 46 research studies of significance. We found that these studies pertained to three main areas of focus concerning the uses of data-driven approaches in game player modeling. One research area involved the objectives of data-driven approaches in game player modeling: behavior modeling and goal recognition. Another concerned methods: classification, clustering, regression, and evolutionary algorithm. The third was comprised of the current challenges and promising research directions for data-driven approaches in game player modeling. [ABSTRACT FROM AUTHOR]

Slicing is a technique, traditionally applied to programs, for extracting the parts of a program that affect the values computed at a statement of interest. In recent years authors have begun to consider slicing at model level. We present a detailed review of existing work on slicing at the level of finite-state-machine-based models. We focus on state-based modeling notations because these have received sufficient attention from the slicing community that there is now a coherent body of hitherto unsurveyed work. We also identify the challenges that state-based slicing presents and how the existing literature has addressed these. We conclude by identifying problems that remain open either because of the challenges involved in addressing them or because the community simply has yet to turn its attention to solving them [ABSTRACT FROM AUTHOR]

A large class of applications need to execute the same workflow on different datasets of identical size. Efficient execution of such applications necessitates intelligent distribution of the application components and tasks on a parallel machine, and the execution can be orchestrated by utilizing task, data, pipelined, and/or replicated parallelism. The scheduling problem that encompasses all of these techniques is called pipelined workflow scheduling, and it has been widely studied in the last decade. Multiple models and algorithms have flourished to tackle various programming paradigms, constraints, machine behaviors, or optimization goals. This article surveys the field by summing up and structuring known results and approaches. [ABSTRACT FROM AUTHOR]

This article reviews four popular mathematical formalisms-queueing theory, network calculus, schedulability analysis, and data flow analysis-and how they have been applied to the analysis of on-chip communication performance in Systems-on-Chip. The article discusses the basic concepts and results of each formalism and provides examples of how they have been used in Networks-on-Chip (NoCs) performance analysis. Also, the respective strengths and weaknesses of each technique and its suitability for a specific purpose are investigated. An open research issue is a unified analytical model for a comprehensive performance evaluation of NoCs. To this end, this article reviews the attempts that have been made to bridge these formalisms. [ABSTRACT FROM AUTHOR]

When attempting to determine how to respond optimally to a large-scale emergency, the ability to predict the consequences of certain courses of action in silico is of great utility. Agent-based simulations (ABSs) have become the de facto tool for this purpose; however, they may be used and implemented in a variety of ways. This article reviews existing implementations of ABSs for large-scale emergency response, and presents a taxonomy classifying them by usage. Opportunities for improving ABS for large-scale emergency response are identified. [ABSTRACT FROM AUTHOR]

The September 01, 1996 issue of the journal "ACM Computing Surveys" contains four articles that span object-oriented programming, partial evaluation, error detection and hardware performance. Writer Antero Taivalasaari's comprehensive paper examines many forms and uses of inheritance in a readable way. He distinguishes among classification, generalization, aggregation and grouping between sub-classing and sub-typing and between specification and implementation inheritance. Writer Neil Jones identifies partial evaluation with automatic program specialization with automatic program specialization and shows that some forms of specialization yield program speedup while others do not. Writer Joseph Gallian focuses on the use of check digits for error detection. A nice feature of that article is emphasis on applications like supermarket product cods and credit card, library, blood bank and money bank applications. Writers Peter Soderquist and Miriam Leeser examines area and performance tradeoffs for floating-point divide and square root implementations. This study combines the presentation of algorithms and hardware/software tradeoffs with performance simulation results.