Winfried Hahn, A. Hagerer, R. Kandlbinder · 2002
MuSiC, the highly-parallel Munich Simulation Computer, represents an approach for hardware-accelerated logic simulation by applying concepts developed for dataflow architectures to high-speed simulation of digital systems. This approach exploiting parallelism inherent in a design is most efficient. In comparison to two different dataflow computation schemes and their hardware-accelerated implementations, this paper shows that the strategy of compiler-driven simulation can be combined with the concept of event-(activity)-directed simulation (selective trace simulation) not only for logic simulation but also for concurrent fault simulation. Experiments indicate that there is a performance advantage of eventflow computing over the algorithmically simpler dataflow computing schemes but the advantage is limited, since dataflow computing performance of a MuSiC version with 256 Processing Units already is in the range of 10/sup 7/ to 10/sup 8/ test-vectors times gates evaluated per second.
Cited by 0
Marshall, Robert Clarence · K-State Research Exchange (Kansas State University) · 2017
Cited by 0
Philippe Matherat, Marc-Thierry Jaekel · ArXiv.org · 2001
Concrete computing machines, either sequential or concurrent, rely on an intimate relation between computation and time. We recall the general characteristic properties of physical time and of present realizations of computing systems. We emphasize the role of computing interferences, i.e. the necessity to avoid them in order to give a causal implementation to logical operations. We compare synchronous and asynchronous systems, and make a brief survey of some methods used to deal with computing interferences. Using a graphic representation, we show that synchronous and asynchronous circuits reflect the same opposition as the Newtonian and relativistic causal structures for physical space-time.
Cited by 5Open access
Philippe Matherat, MARC-THIERRY JAEKEL · Mathematical Structures in Computer Science · 2003
Concrete computing machines, either sequential or concurrent, rely on an intimate relationship between computation and time. We recall the general characteristic properties of physical time and of present realisations of computing systems. We emphasise the role of computing interferences, that is, the necessity of avoiding them in order to give a causal implementation to logical operations. We compare synchronous and asynchronous systems, and give a brief survey of some methods used to deal with computing interferences. Using a graphic representation, we show that synchronous and asynchronous circuits reflect the same opposition as the Newtonian and relativistic causal structures for physical space-time.
Cited by 4
Vaidy Sunderam · 1994
this paper, we discuss critical design and implementation issues in heterogeneous concurrent computing, and describe techniques for enhancing its effectiveness. In particular, we highlight the system level infrastructures that are required, aspects of parallel algorithm development that most affect performance, system capabilities and limitations, and tools and methodologies for effective computing in heterogeneous networked environments. We also present recent developments and experiences in the context of the PVM system, and comment on ongoing and future work. 1. INTRODUCTION
Cited by 2
Adam J. Ferrari, Vaidy Sunderam · 2002
The TPVM (Threads-oriented PVM) system, is an experimental auxiliary subsystem for the PVM distributed system, which supports the use of lightweight processes or "threads" as the basic unit of parallelism and scheduling. TPVM provides a library interface which presents both a traditional, task based, explicit message passing model, as well as a data-driven scheduling model that enables straightforward specification of computation based on data dependencies. Our system design is still under development, but a prototype implementation has allowed us to perform a number of preliminary experiments. These have provided strong evidence that TPVM can offer improved performance, processor utilization, and load balance to several application categories. Through our experiments we have also determined that the current TPVM design is not very well suited to certain types of applications, most notably highly synchronous, SPMD-style algorithms.
Cited by 34
Matherat, Philippe, Jaekel, Marc-Thierry · arXiv (Cornell University) · 2001
Concrete computing machines, either sequential or concurrent, rely on an intimate relation between computation and time. We recall the general characteristic properties of physical time and of present realizations of computing systems. We emphasize the role of computing interferences, i.e. the necessity to avoid them in order to give a causal implementation to logical operations. We compare synchronous and asynchronous systems, and make a brief survey of some methods used to deal with computing interferences. Using a graphic representation, we show that synchronous and asynchronous circuits reflect the same opposition as the Newtonian and relativistic causal structures for physical space-time.
Cited by 0Open access
Microprocessors and Microsystems · 1987
Cited by 1
Daniel Gebler, Vahid Hashemi, Andrea Turrini · Lecture notes in computer science · 2014
Cited by 7
Virginia Klema · 1986
Abstract : An important issue in harmonic retrieval problems, as in most computational problems, is the well-posedness of a particular problem instance. Specifically, when is a harmonic signal reconstruction by these methods particularly densitive to additive noise? A definitive answer to this question was obtained in the form of a condition number for harmonic retrieval problems and is based on the Vandermonde determinant of the harmonic frequencies. This result was obtained via a combination of relationships using classical trigonometric moment theory and Toeplitz matrix conditioning. A fundamental new result, due to Davis and Bhatia, on the spectral sensitivity of unitary matrices was used. The result first appeared in the Princeton Conference on Information Systems and Sciences Proceedings in a paper titled Conditioning of Eigenvector Methods for Beamforming Problems while more recently a survey of this work will form the basis for an invited presentation at the upcoming IEEE Workshop on Spectrum Estimation. That survey is titled The Sensitivity of Beamforming Problems.
Cited by 0
Jaya Narain · Spectrum Research Repository (Concordia University) · 1998
Parallel computing is an intricate mix of marketplace requirements, architectural understanding, technology issues, and issues concerning costs. One controlling myth is that high-volume commodity processors must, by the nature of things, be the common building blocks for both desktop clients and room-size servers. This myth--and the supporting myth of architectural convergence of clients and servers--should be subject to dispassionate analysis. With only one program counter per processor, conventional processors are becoming increasingly unresponsive in spite of faster clock rates. We will show that reliance on Instruction-Level Parallelism (ILP) for performance drives processor state upwards. When this massive state is not distributed across multiple program counters, processors choke on their own expensive context switches, here reconceptualized to show their true cost. Within the framework of Little's law from queueing theory, we analyze conventional RISC superscalar processors as a case study of the inadequacy of the class of "ILP" processors. We contrast this to multithreaded processors that exploit both ILP and Thread-Level Parallelism (TLP). As a contribution to parallel programming, we show how data caches in multithreaded architectures can be used to manage speculative state, and perform atomic updates involving multiple variables. There is no "convergence architecture"; there are only divergence architectures
Cited by 0Open access
John U. Duselis, Isaac D. Scherson · 2009
We introduce a computational framework for concurrent adaptive computing in heterogeneous environments for computationally intensive applications. This framework considers the presence of inter-connected computational resources which are discoverable and a workload which needs to be executed either by concurrent means or on a singular resource. The selection of resources, using a novel measurement of performance, leads to the adaptive inclusion/exclusion of resources to be used in the efficient execution of workload computations. The adaptive approach is that it makes a determination to include a proper subset of resources for system inclusion to execute a workload, which is contrary (except when the subset is all-inclusive) to a greedy approach where all the resources are seized for the workload application. The selection of a subset of resources may be more efficient due to the high level of heterogeneity of the resources, where, for some resources, certain resource selections may be detrimental or have no value to send work there. Furthermore, this framework aims to lessen the unpredictability and uncontrollability of heterogeneous systems by using this analysis for resource selection.
Cited by 2
Moritz Wundke · The Oberta in Open Access (Universitat Oberta de Catalunya) · 2014
Concurrent aims to be a different type of task distribution system compared to what MPI like system do. It adds a simple but powerful application abstraction layer to distribute the logic of an entire application onto a swarm of clusters holding similarities with volunteer computing systems. Traditional task distributed systems will just perform simple tasks onto the distributed system and wait for results. Concurrent goes one step further by letting the tasks and the application decide what to do. The programming paradigm is then totally async without any waits for results and based on notifications once a computation has been performed.
Cited by 0Open access
Laura Waller · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 1986
An experimental chip that incorporates a novel message-routing scheme may hold the key to the second generation of concurrent computers, according to Charles L. Seitz, a pioneer in this form of computing. A professor of computer science at the California Institute of Technology and developer of the Cosmic Cube Architecture for Linking Multiple Processors. Seitz worked with doctoral candidate William J. Dally, who designed and fabricated the Torus Routing Chip - named for its doughnut-shaped topology. Using a scheme similar to virtual computer memory, the chips seeks to overcome a hang-up caused by the very feature that gives concurrent computing schemes their boosted horsepower: large numbers of microprocessor nodes working simultaneously on the same task. Messages running between nodes can get logjammed in single-dimension communication channels, thus deadlocking operation.
Cited by 0
Virginia Klema · 1990
Cited by 0
D. M. Kalai Selvi, V. Ceronmani Sharmila, P. Ezhumalai et al. · 2025
By enabling sectors can purchase or rent cloud-based service the providers computer, preservation and network infrastructure. The cloud-based service providers, the cloud computing paradigm is enabling creative and disruptive services (for example, the IBM and Google cloud platforms). Virtual network embedding (VNE) is the process of adding a virtual network to an already existing physical network. Cloud service providers (CSPs) confront several important issues, including energy usage, service level agreement violations, unexpected services failures that directly affect Qos, and significant revenue loss. The cloud computing fat-tree data center network uses an optimal apparent-integrated virtual network embedding technique to get around this. The dispersed and aggregated integration of the machine simulation into several physical machines is what makes the suggested algorithm proprietary information valuable. In simpler terms, the suggested technique is implemented across several physical servers to map the virtual network's virtual infrastructure simultaneously, decreasing the time required for mapping of resources while simultaneously improving the quality of service (QoS) offered by CSO
Cited by 0
Jianjun Zhao, J. Cheng, Kentaro Ushijima · 2002
Program slicing has many applications in software engineering activities. However, until recently, no slicing algorithm has been presented that can compute executable slices for concurrent logic programs. We present a dependence-graph based approach to computing executable slices for concurrent logic programs. The dependence-based representation used in the paper is called the Argument Dependence Net which can be used to explicitly represent various types of program dependences in a concurrent logic program. Based on the ADN, we can compute static executable slices for concurrent logic programs at argument level.
Cited by 7
F. Warren Burton · The Computer Journal · 1987
There are at least two approaches to the design of languages for parallel computing. One approach is to use functional or relational languages which are easy to read, write, transform and verify. The more conventional approach is to use procedural languages which give a programmer a high degree of control over the run-time behaviour of a program. There is a need to reconcile these two approaches in a language which permits both simplicity and efficiency. We propose a small and simple set of annotations (or pragmas) to control the run-time behaviour of a functional program. The annotations allow a programmer to use three forms of parameter passing. The parameter-passing mechanisms correspond to passing by name, value and need in a sequential language. In addition, in a distributed system a programmer can specify that work should be done on the current processor, an arbitrary processor, or a particular processor such as the one containing a specific data item. The annotations cannot affect the meaning (result) of a functional program, except for causing non-termination in some cases (which we view as an extreme form of inefficiency). This separation of meaning from control allows a program to be both simple and efficient. Since non-determinism appears to be unavoidable without significant loss of efficiency in a concurrent system, the interaction of the proposed annotations with non-determinism is briefly considered. The run-time behaviour of an annotated functional program is similar to that of procedural programs using message passing, semaphores or rendezvous to control communication and synchronisation.
Cited by 21Open access
Paul A. Gray, Vaidy Sunderam · Journal of Systems Architecture · 1999
Cited by 0
柴山悦哉, ETSUYA SHIBAYAMA · Tokyo Tech Research Repository (Tokyo Institute of Technology) · 2006
Cited by 0