检索结果-内蒙古大学图书馆

10th International Symposium on Pervasive Systems, Algorithms, and Networks

作者： Luo, Yang Lam, King Tin Wang, Cho-Li Univ Hong Kong Dept Comp Sci Hong Kong Hong Kong Peoples R China

ISBN: (纸本)9781424454037

This paper presents our studies on the connectivity between objects and traversal behavior over the access paths among objects in order to devise profitable prefetching policies for object-based distributed systems. We propose a profiling strategy that can classify classes and fields into a handful of generic types exhibiting distinctive and exploitable access patterns during the runtime. Based on the classifications, we propose an improved algorithm of object prefetching to select best candidates to prefetch under practical message size limits. We implement the methodology into our JESSICA2 distributed Java virtual machine and evaluate its effectiveness. Our experimental results show that our prefetching policies are able to eliminate over 93% cache coherence protocol messages and halve the execution time for fine-grained applications.

关键词： object sharing prefetching object access patterns distributed Java virtual machine distributed shared memory

来源：评论

学校读者我要写书评

暂无评论

tinyDSM: A Highly Reliable Cooperative Data Storage For Wireless Sensor Networks

tinyDSM: A Highly Reliable Cooperative Data Storage For Wire...

引用

International Symposium on Collaborative Technologies and Systems

作者： Piotrowski, Krzysztof Langendoerfer, Peter Peter, Steffen IHP D-15236 Frankfurt Germany

ISBN: (纸本)9781424445851

The advantage of a Wireless Sensor Network (WSN) compared to a centric approach is the distribution of sensing suites. However in order for such a system of distributed resources to work in a reliable and effective way a smart cooperation between nodes is needed. In this paper we propose a middleware approach for a highly reliable data storage that helps to assure data availability despite the well known WSN resource problems and disappearing or inactive nodes by providing a reasonable data redundancy in the system. Such a solution helps to ease the design and optimization of the data exchange between nodes as well. Our solution is configurable in order to satisfy the needs of the application on top regarding performance/requirements trade-off. The options specify the quantity and quality of the data replication. Additional features like event mechanism that monitors the data and the possibility to issue database like queries increase the applicability of our middleware. In this paper we focus on the evaluation of its capabilities regarding reliability, the consistency of replicates and the costs of the data management. The simulation results for a reasonable set-up show that the CPU load caused by the data replication is low (below 3 percent) and the average inconsistency time is as small as about 0,06 seconds for a single hop and about 0, 15 seconds for a two hops replication area. There is still room for improvements, but a clear definition of problems helps to find ways to cope with them in order to achieve the chosen goals.

关键词： Cooperative Computing Wireless Sensor Networks Software Development distributed shared memory distributed Data Storage

来源：评论

学校读者我要写书评

暂无评论

Failure resilient real-time data federation system 09

Failure resilient real-time data federation system

引用

Proceedings of the 2009 Spring Simulation Multiconference

作者： Aakanksha Gagrani Brijesh Pillai Srikumar Krishnamoorthy Infosys Technologies Pvt. Ltd.

Data federation systems virtualize access to enterprize data resources by integrating data from disparate and heterogeneous operational data sources in an on-demand and real-time basis. The key challenge of low latency data access in such real-time data federation systems can be addressed by grid based scale-out architecture. However, failure of resources in the grid can pose serious challenges in data federation as the query processing is federated over multiple grid nodes. In such real-time data federation systems, it is often desirable to recover from failure and continue operation rather than repeat the entire process. This paper proposes a decentralized failure-recovery protocol for data federation system using data spaces based architecture. The generic nature of the protocol makes it extensible to applications other than data federation system as well. Moreover, the protocol does not make any assumptions about the availability of any central repository for recovering from failure. We implement the proposed failure recovery protocol in a simulation environment and present the key findings of the study.

关键词： failover grid computing distributed query processing distributed shared memory fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Moving address translation closer to memory in distributed shared-memory multiprocessors

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2005年第7期16卷 612-623页

作者： Qiu, XG Dubois, M Sun Microsyst Inc Palo Alto CA 94303 USA Univ So Calif Dept Elect Engn Syst Los Angeles CA 90089 USA

To support a global virtual memory space, an architecture must translate virtual addresses dynamically. In current processors, the translation is done in a TLB (Translation Lookaside Buffer), before or in parallel with the first-level cache access. As processor technology improves at a rapid pace and the working sets of new applications grow insatiably, the latency and bandwidth demands on the TLB are difficult to meet, especially in multiprocessor systems, which run larger applications and are plagued by the TLB consistency problem. We describe and compare five options for virtual address translation in the context of distributed shared memory (DSM) multiprocessors, including CC-NUMAs (Cache-Coherent Non-Uniform memory Access Architectures) and COMAs ( Cache Only memory Access Architectures). In CC-NUMAs, moving the TLB to shared memory is a bad idea because page placement, migration, and replication are all constrained by the virtual page address, which greatly affects processor node access locality. In the context of COMAs, the allocation of pages to processor nodes is not as critical because memory blocks can dynamically migrate and replicate freely among nodes. As the address translation is done deeper in the memory hierarchy, the frequency of translations drops because of the filtering effect. We also observe that the TLB is very effective when it is merged with the shared-memory, because of the sharing and prefetching effects and because there is no need to maintain TLB consistency. Even if the effectiveness of the TLB merged with the shared memory is very high, we also show that the TLB can be removed in a system with address translation done in memory because the frequency of translations is very low.

关键词： multiprocessors distributed shared memory virtual memory simulations dynamic address translation virtual-address caches

来源：评论

学校读者我要写书评

暂无评论

A parametrized algorithm that implements sequential, causal, and cache memory consistencies

引用

JOURNAL OF SYSTEMS AND SOFTWARE 2008年第1期81卷 120-131页

作者： Jimenez, Ernesto Fernandez, Antonio Cholvi, Vicent Univ Jaume 1 Dept Lenguajes & Sist Informat Castellon de La Plana 12071 Spain Univ Politecn Madrid Madrid 28031 Spain Univ Rey Juan Carlos Mostoles 28933 Spain

In this paper, we present an algorithm that can be used to implement sequential, causal, or cache consistency in distributed shared memory (DSM) systems. For this purpose it includes a parameter that allows us to choose the consistency model to be implemented. If all processes run the algorithm with the same value in this parameter, the corresponding consistency is achieved. (Additionally, the algorithm tolerates that processes use certain combination of parameter values.) This characteristic allows a concrete consistency model to be chosen, but implements it with the more efficient algorithm in each case (depending of the requirements of the applications). Additionally, as far as we know, this is the first algorithm proposed that implements cache coherence. In our algorithm, all the read and write operations are executed locally when implementing causal and cache consistency (i.e., they are fast). It is known that no sequential algorithm has only fast memory operations. In our algorithm, however, all the write operations and some read operations are fast when implementing sequential consistency. The algorithm uses propagation and full replication, where the values written by a process are propagated to the rest of the processes. It works in a cyclic turn fashion, with each process of the DSM system, broadcasting one message in turn. The values written by the process are sent in the message (instead of sending one message for each write operation): However, unnecessary values are excluded. All this permits the amount of message traffic owing to the algorithm to be controlled. (c) 2007 Elsevier Inc. All rights reserved.

关键词： distributed shared memory causal consistency sequential consistency distributed systems coherency models

来源：评论

学校读者我要写书评

暂无评论

The read-copy-update mechanism for supporting real-time applications on shared-memory multiprocessor systems with Linux

引用

IBM SYSTEMS JOURNAL 2008年第2期47卷 221-236页

作者： Guniguntala, D. McKenney, P. E. Triplett, J. Walpole, J. IBM India Syst & Technol Lab Bangalore 560071 Karnataka India Linux Technol Ctr Bangalore Karnataka India IBM Beaverton Lab Beaverton OR 97006 USA Portland State Univ Hillsboro OR 97124 USA Portland State Univ Beaverton OR 97007 USA

Read-copy update (RCU) is a synchronization mechanism in the Linux (TM) kernel that provides significant improvements in multiprocessor scalability by eliminating the writer-delay problem of readers-writer locking. RCU implementations to date, however, have had the side effect of expanding non-preemptible regions of code, thereby degrading real-time response. We present here a variant of RCU that allows preemption of read-side critical sections and thus is better suited for real-time applications. We summarize priority-inversion issues with locking, present an overview of the RCU mechanism, discuss our counter-based adaptation of RCU for real-time use, describe an additional adaptation of RCU that permits general blocking in readside critical sections, and present performance results. We also discuss an approach for replacing the readers-writer synchronization with RCU in existing implementations.

关键词： MULTIPROCESSORS SCALABILITY (Systems engineering) REAL-time control distributed shared memory LINUX operating systems SYNCHRONIZATION KERNEL functions

来源：评论

学校读者我要写书评

暂无评论

On the effectiveness of runtime techniques to reduce memory sharing overheads in distributed Java implementations

引用

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE 2008年第13期20卷 1509-1538页

作者： Lobosco, Marcelo Loques, Orlando de Amorim, Claudio L. Univ Fed Juiz de Fora Dept Ciencia Comp Juiz De For A MG Brazil Univ Fed Fluminense Inst Comp BR-24210240 Niteroi RJ Brazil Univ Fed Rio de Janeiro COPPE PESC Lab Comp Paralela Rio de Janeiro Brazil

distributed Java virtual machine (dJVM) systems enable concurrent Java applications to transparently run on clusters of commodity computers. This is achieved by supporting Java's shared-memory model over multiple JVMs distributed across the cluster's computer nodes. In this work, we describe and evaluate selective dynatnic diffing and lazy home allocation, two new runtime techniques that enable dJVMs to efficiently support memory sharing across the cluster. Specifically, the two proposed techniques can contribute to reduce the overheads due to message traffic, extra memory space, and high latency of remote memory accesses that such dJVM systems require for implementing their memory-coherence protocol either in isolation or in combination. In order to evaluate the performance-related benefits of dynamic diffing and lazy home allocation, we implemented both techniques in Cooperative JVM (CoJVM), it basic dJVM system we developed in previous work. In subsequent work, we carried out performance comparisons between the basic CoJVM and modified CoJVM versions for five representative concurrent lava applications (matrix multiply, LU, Radix, fast Fourier transform, and SOR) using our proposed techniques. Our experimental results showed that dynamic doffing and lazy home allocation significantly reduced memory sharing overheads. The reduction resulted in considerable gains in CoJVM system's performance, ranging front 9% tip to 20%, in four out of the five applications, with resulting speedups varying front 6.5 tip to 8.1 for in 8-node cluster of computers. Copyright (C) 2007 John Wiley & Sons, Ltd.

关键词： Java JVM distributed shared memory cluster computing concurrent Java applications high performance computing

来源：评论

学校读者我要写书评

暂无评论

A wireless sensor network reliable architecture for intrusion detection

A wireless sensor network reliable architecture for intrusio...

引用

4th Conference on Next Generation Internet Networks

作者： Lanuendorfer, Peter Grilo, Antonio, Jr. Piotrowski, Krzysztof Casaca, August IHP Technol Pk 25 D-15236 Frankfurt Oder Germany IST INESC P-1000029 Lisbon Portugal

ISBN: (纸本)9781424417841

An important aspect of homeland security is securing public buildings and places against intrusion. Wireless sensor networks (WSN) are an efficient means of detecting intrusion and extremely convenient to deploy in such situations. However, the software running on top of the sensor nodes that directly supports the alarm applications defined by the security forces needs to provide reliability, security and ease of use. In this paper we introduce the new concept of tinyDSM that combines distributed shared memory like behaviour with event mechanisms and a new transport protocol called DTSN which provides reliable and efficient transport of messages in WSN. Both contribute to the WSN reliability that is required for the intrusion detection application. The specific security aspects are not treated in this paper.

关键词： wireless sensor networks homeland security intrusion detection distributed shared memory transport protocols

来源：评论

学校读者我要写书评

暂无评论

A page-coherent, causally consistent protocol for distributed shared memory

引用

JOURNAL OF SYSTEMS AND SOFTWARE 2004年第3期72卷 305-319页

作者： Campos, AE Navarro, JE Pontificia Univ Catolica Chile Dept Ciencia Computac Santiago 22 Chile

distributed shared memory (DSM) provides a virtual address space that is shared among processors in a distributed system. It allows application programmers to elude message passing and use the more familiar shared-memory paradigm. To increase efficiency, DSM implementations replicate memory pages, introducing the problem of consistency. As fewer restrictions are imposed to the replicas, more efficient implementations are possible, but application programming becomes more difficult. Causal consistency is a model that offers a balance, by allowing efficient implementations without significantly increasing programming difficulty. This work presents a page-coherent, causally consistent DSM protocol. This protocol requires that each time a node sends the contents of a page p, it piggybacks a list of all pages the receiver must invalidate. This list includes all pages written by operations unknown to the receiver, and that potentially causally precede the most recent write operation over p known by the sender. The protocol's main feature is less communication overhead than other causally consistent protocols in the literature. This behavior was confirmed running test applications on top of a simple simulator. (C) 2003 Elsevier Inc. All rights reserved.

关键词： distributed computing Multiprocessors Computer science Computer software distributed shared memory memory management (Computer science)

来源：评论

学校读者我要写书评

暂无评论

RAPID: Reconfigurable and scalable all-photonic interconnect for distributed shared memory multiprocessors

引用

JOURNAL OF LIGHTWAVE TECHNOLOGY 2004年第9期22卷 2101-2110页

作者： Kodi, AK Louri, A Univ Arizona Dept Elect & Comp Engn Tucson AZ 85721 USA

In this paper, we describe the design and analysis of a scalable architecture suitable for large-scale distributed shared memory (DSM) systems. The approach is based on an interconnect technology which combines optical components and a novel architecture design. In DSM systems, numerous shared memory transactions such as requests, responses and acknowledgment messages propagate simultaneously in the network. As the network size increases, network contention results in increasing the critical remote memory access latency, which significantly penalizes the performance of DSM systems. In our proposed architecture called reconfigurable and scalable all-photonic interconnect for distributed-shared memory (RAPID), we provide high connectivity by maximizing the channel availability for remote communication to reduce the critical remote latency. RAPID provides fast and efficient unicast, multicast and broadcast capabilities using a combination of aggressively designed wavelength division multiplexing (WDM), time division multiplexing (TDM), and space division multiplexing (SDM) techniques. RAPID is wavelength-routed, permitting the same limited set of wavelength to be reused among all processors., We evaluated RAPID based on network characteristics, power budget criteria, and by simulation using synthetic traffic workloads and compared it against other networks such as electrical ring, torus, mesh, and hypercube networks. We found that RAPID outperforms all networks and still provides good performance as the network is scaled to very large numbers.

关键词： cache coherence distributed shared memory optical interconnects scalable optical networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：