Search results for "Distributed computing"

Modeling and Simulation of Network-on-Chip Systems with DEVS and DEUS

2013

Networks on-chip (NoCs) provide enhanced performance, scalability, modularity, and design productivity as compared with previous communication architectures for VLSI systems on-chip (SoCs), such as buses and dedicated signal wires. Since the NoC design space is very large and high dimensional, evaluation methodologies rely heavily on analytical modeling and simulation. Unfortunately, there is no standard modeling framework. In this paper we illustrate how to design and evaluate NoCs by integrating the Discrete Event System Specification (DEVS) modeling framework and the simulation environment called DEUS. The advantage of such an approach is that both DEVS and DEUS support modularity—the fo…

Hierarchies of probabilistic and team FIN-learning

2001

AbstractA FIN-learning machine M receives successive values of the function f it is learning and at some moment outputs a conjecture which should be a correct index of f. FIN learning has two extensions: (1) If M flips fair coins and learns a function with certain probability p, we have FIN〈p〉-learning. (2) When n machines simultaneously try to learn the same function f and at least k of these machines output correct indices of f, we have learning by a [k,n]FIN team. Sometimes a team or a probabilistic learner can simulate another one, if their probabilities p1,p2 (or team success ratios k1/n1,k2/n2) are close enough (Daley et al., in: Valiant, Waranth (Eds.), Proc. 5th Annual Workshop on C…

Balls into non-uniform bins

2014

Balls-into-bins games for uniform bins are widely used to model randomized load balancing strategies. Recently, balls-into-bins games have been analysed under the assumption that the selection probabilities for bins are not uniformly distributed. These new models are motivated by properties of many peer-to-peer (P2P) networks, which are not able to perfectly balance the load over the bins. While previous evaluations try to find strategies for uniform bins under non-uniform bin selection probabilities, this paper investigates heterogeneous bins, where the "capacities" of the bins might differ significantly. We show that heterogeneous environments can even help to distribute the load more eve…

Online Scheduling of Task Graphs on Heterogeneous Platforms

2020

Modern computing platforms commonly include accelerators. We target the problem of scheduling applications modeled as task graphs on hybrid platforms made of two types of resources, such as CPUs and GPUs. We consider that task graphs are uncovered dynamically, and that the scheduler has information only on the available tasks, i.e., tasks whose predecessors have all been completed. Each task can be processed by either a CPU or a GPU, and the corresponding processing times are known. Our study extends a previous $4\sqrt{m/k}$ 4 m / k -competitive online algorithm by Amaris et al. [1] , where $m$ m is the number of CPUs and $k$ k the number of GPUs ( $m\geq k$ m ≥ k ). We prove that no online…

Semantics for Managing Systems in Heterogeneous and Distributed Environment

2002

The problem of legacy systems collaboration is being solved. Particularly we look at the collaboration as workflow in a distributed and heterogeneous environment. Attention is paid to the description of semantics for workflow process definition languages. There are many solutions how semantics can be decomposed into logical fragments, but the problem of obtaining reusable components that are easy to compile into desired specific semantics still remains. We evolve the division of semantics by semantic aspects whose description is based on abstract data types (pre-built components) and connectors (meta-programs to produce the glue code) between them. This paper offers a way in which semantic …

Use of Supercomputing towards the Generation of a Process Execution Plan in Distributed Real-Time Systems

2010

This work consider the scheduling of periodic tasks or processes with real-time constraints in a distributed environment. Each task must be executed meeting deadlines, precedence relationships and resources constraints. The problem of scheduling tasks on a distributed environment can be viewed as a problem of assigning processes to the processors but keeping the schedulability in local environments. Thus, the problem can be divided in two phases: the first phase is assigning processes to processors and the second is to schedule assigned processes in each processor in the distributed environment. This paper focuses in the first phase. It introduces a heuristic mechanism for assigning process…

Potential implementation of reservoir computing models based on magnetic skyrmions

2018

Reservoir Computing is a type of recursive neural network commonly used for recognizing and predicting spatio-temporal events relying on a complex hierarchy of nested feedback loops to generate a memory functionality. The Reservoir Computing paradigm does not require any knowledge of the reservoir topology or node weights for training purposes and can therefore utilize naturally existing networks formed by a wide variety of physical processes. Most efforts prior to this have focused on utilizing memristor techniques to implement recursive neural networks. This paper examines the potential of skyrmion fabrics formed in magnets with broken inversion symmetry that may provide an attractive phy…

Object Clustering Methods and a Query Decomposition Strategy for Distributed Object-Based Information Systems

1999

Emerging developments and advances in distributed processing have created a need for tools and methods to partition and distribute information systems across interconnected processors. In particular, distribution approaches which take into account the key characteristics of OO concepts are required to extend traditional fragmentation results to object oriented database systems. To fulfill the above requirements, we propose a methodology for the distribution design of object-based information systems. The underlying approach consists of techniques and heuristics that can be used to create clusters of inter-related object classes that can be fragmented interdependently, producing distribution…

HybridS: A Scheme for Secure Distributed Data Storage in WSNs

2008

In unattended wireless sensor networks (WSNs), data is stored locally or at designated nodes upon sensing, and users can access it on demand. This paradigm can improve energy efficiency by making use of the upcoming cheap and large flash memory, as well as system robustness. Nevertheless, the security and dependability of distributed storage are critical for the applicability of such WSNs. In this paper, we propose a secure and dependable data storage scheme by taking advantages of secret sharing and Reed-Solomon code, which has computational security and yet maintains optimal data size. The extensive analysis verifies our scheme can provide secure and dependable data storage in WSNs in the…

A recognize-and-accuse policy to speed up distributed processes

1994