Powerlists are recursive data structures that together with their associated algebraic theories could offer both a methodology to design parallel algorithms and parallel programming abstractions to ease the developmen...
详细信息
ISBN:
(纸本)9781728116440
Powerlists are recursive data structures that together with their associated algebraic theories could offer both a methodology to design parallel algorithms and parallel programming abstractions to ease the development of parallel applications This has been also proved by a concrete development of such a framework that allows easy, efficient, and reliable implementation of Java parallel programs on shared memory systems. The paper presents a highly scalable version of this framework by extending it to distributed memory systems based on an MPI implementation. Through this extension we may use the framework to develop Java parallel programs also on distributed memory systems such as clusters. The design of the framework enables flexibility in defining the appropriate execution type depending on the execution system and its characteristics. Therefore, it is possible to choose MPI execution (ithat also could be combined with multithreading) if the available system includes an MPI platform, or simple multithreading execution. Examples are given and performance experiments are conducted. The performance analysis of these applications emphasises the utility and the efficiency of this framework extension.
This article reviews record versions processing (reconciliation). It provides an algorithm for NoSQL database record versions management. It develops a record versions reconciliation model that is simultaneously updat...
详细信息
ISBN:
(纸本)9781509060580
This article reviews record versions processing (reconciliation). It provides an algorithm for NoSQL database record versions management. It develops a record versions reconciliation model that is simultaneously updated by multiple users. The model estimates record versions reconciliation time and the number of versions simultaneously stored in a database. based on the modeling results a designer can provide recommendations on the maximum number of users (or applications) simultaneously working with one document (database record). This is important in case there are limitations imposed on the document reconciliation time. It describes preparation and execution of experiments in a cloud, for model adequacy analysis.
parallel communication plays a critical role in massively parallel systems, especially in distributed memory systems executing parallel programs on shared data. Therefore, integrating an interconnection network in the...
详细信息
ISBN:
(纸本)9781467387767
parallel communication plays a critical role in massively parallel systems, especially in distributed memory systems executing parallel programs on shared data. Therefore, integrating an interconnection network in these systems becomes essential to ensure data inter-nodes exchange. Choose the most effective communication structure must meet certain criteria: speed, size and power consumption. Indeed, the communication phase should be as fast as possible to avoid compromising parallel computing, using small and low power consumption modules to facilitate the interconnection network extensibility in a scalable system. To meet these criteria and based on a module reuse methodology, we chose to integrate a reconfigurable SCAC-Net interconnection network to communicate data in SCAC Massively parallel SoC. This paper presents the detailed hardware implementation and discusses the performance evaluation of the proposed reconfigurable SCAC-Net network.
A very ambitious objective in the field of policy-based systems is the provision of an intuitive and transparent way for policy specification, refinement and enforcement. This is one of the key enabling technologies f...
详细信息
ISBN:
(纸本)9780769539393
A very ambitious objective in the field of policy-based systems is the provision of an intuitive and transparent way for policy specification, refinement and enforcement. This is one of the key enabling technologies for a simplified security management of complex networked environments. Currently, security policies are enforced by configuring the end devices by means of low-level device-specific parameters manually derived from high level specifications. This process, defined as policy translation, is still performed without a holistic view of the overall security requirements. This paper presents the network Contextualization Tool (NCTool), a software supporting administrators in performing network dependent activities when configuring security enabled devices. The tool provides a great advantage in the management of complex networks. In fact, it simplifies the network administration tasks and reduces effort and responsibilities for the administrators, thus decreasing the risk of mistaken configurations.
The paper considers an integrated proactive framework for defense against spreading network worms in the Internet. The framework is intended for network worm detection (by recognizing the actions on scanning of networ...
详细信息
ISBN:
(纸本)9780769535449
The paper considers an integrated proactive framework for defense against spreading network worms in the Internet. The framework is intended for network worm detection (by recognizing the actions on scanning of network hosts) and containment of worm spreading (by limiting and blocking the packets transmitted by infected hosts). The framework is based on application of different heuristic detection and response mechanisms, their combination and automatic dynamic adaptation according to current network conditions. The paper describes the software system for simulation and evaluation of defense mechanisms investigated against spreading network worms and the results of experiments on detection and containment of network worms.
parallelprocessing architectures have been increasingly utilized due to their potential for improving performance and energy efficiency. Unfortunately, the anticipated improvement often suffers from a limitation caus...
详细信息
ISBN:
(纸本)9781467387767
parallelprocessing architectures have been increasingly utilized due to their potential for improving performance and energy efficiency. Unfortunately, the anticipated improvement often suffers from a limitation caused by memory access latency and latency variation, which consequently impact Quality of Service (QoS). This paper presents a service guaranteed multi-port packet memory system to boost parallelism in protocol processing architectures. In this proposed memory system, all arriving packets are guaranteed a memory space, such that, a packet memory space can be allocated in a bounded number of cycles and each of its locations is accessible in a single cycle. We consider a real-time Voice Over Internet Protocol (VOIP) call as a case-study to evaluate our service guaranteed memory system.
This paper presents the design of a P2P data persistent platform. Durable access and integrity of the data are ensured despite massive attacks. This platform, named DataCube, exploits the properties of cluster-based p...
详细信息
ISBN:
(纸本)9780769539393
This paper presents the design of a P2P data persistent platform. Durable access and integrity of the data are ensured despite massive attacks. This platform, named DataCube, exploits the properties of cluster-based peer-to-peer substrates to implement a compound of full replication and rateless erasure codes. DataCube guarantees durable access and integrity of data despite adversarial attacks. In particular, the recovery of damaged data is achieved through the retrieval of coded blocks whose integrity is checked on the fly.
MPI is the de facto communication standard library for parallel applications in distributed memory architectures. Collective operations performance is critical in HPC applications as they can become the bottleneck of ...
详细信息
ISBN:
(纸本)9781665469586
MPI is the de facto communication standard library for parallel applications in distributed memory architectures. Collective operations performance is critical in HPC applications as they can become the bottleneck of their executions. The advent of larger node sizes on multicore clusters has motivated the exploration of hierarchical collective algorithms aware of the process placement in the cluster and the memory hierarchy. This work analyses and compares several hierarchical collective algorithms from the literature that do not form part of the current MPI standard. We implement the algorithms on top of OpenMPI using the shared-memory facility provided by MPI-3 at the intra-node level and evaluate them on ARM-based multicore clusters. From our results, we evidence aspects of the algorithms that impact the performance and applicability of the different algorithms. Finally, we propose a model that helps us to analyze the scalability of the algorithms.
This work presents a novel approach to distributed training of deep neural networks (DNNs) that aims to overcome the issues related to mainstream approaches to data parallel training. Established techniques for data p...
详细信息
ISBN:
(纸本)9781728116440
This work presents a novel approach to distributed training of deep neural networks (DNNs) that aims to overcome the issues related to mainstream approaches to data parallel training. Established techniques for data parallel training are discussed from both a parallel computing and deep learning perspective, then a different approach is presented that is meant to allow DNN training to scale while retaining good convergence properties. Moreover, an experimental implementation is presented as well as some preliminary results.
This paper presents a novel extension to the Consensus-based Bundle Algorithm (CBBA), which we have named Cluster-Formed Consensus-based Bundle Algorithm (CF-CBBA). CF-CBBA is designed to reduce the amount of communic...
详细信息
ISBN:
(纸本)9781479927289
This paper presents a novel extension to the Consensus-based Bundle Algorithm (CBBA), which we have named Cluster-Formed Consensus-based Bundle Algorithm (CF-CBBA). CF-CBBA is designed to reduce the amount of communication required to complete a distributed task allocation process, by partitioning the problem and processing it in parallel clusters. CF-CBBA has been shown, in comparison with baseline CBBA, to require less communication when allocating tasks. Three key aspects of task allocation have been investigated;(a) the time taken to allocate tasks, (b) the amount of communication necessary to satisfy the requirements of distributed task allocation algorithms such as CBBA, and (c) the efficiency with which a collection of tasks (a mission) is completed by a group of robots (a collective).
暂无评论