检索结果-内蒙古大学图书馆

memory sharing for interactive ray tracing on clusters

PARALLEL COMPUTING 2005年第2期31卷 221-242页

作者： DeMarle, DE Gribble, CP Boulos, S Parker, SG Univ Utah Sci Comp & Imaging Inst Salt Lake City UT 84112 USA

We present recent results in the application of distributed shared memory to image parallel ray tracing on clusters. Image parallel rendering is traditionally limited to scenes that are small enough to be replicated in the memory of each node, because any processor may require access to any piece of the scene. We solve this problem by making all of a cluster's memory available through software distributed shared memory layers. With gigabit ethernet connections, this mechanism is sufficiently fast for interactive rendering of multi-gigabyte datasets. Object- and page-based distributed shared memories are compared, and optimizations for efficient memory use are discussed. (c) 2005 Elsevier B.V. All rights reserved.

关键词： scientific visualization out-of-core rendering distributed shared memory ray tracing cache miss reduction

来源：评论

学校读者我要写书评

暂无评论

Performance evaluation of view-oriented parallel programming

Performance evaluation of view-oriented parallel programming

引用

34th International Conference on Parallel Processing (ICPP)

作者： Huang, Z Purvis, M Werstein, P Univ Otago Dept Comp Sci Dunedin New Zealand

ISBN: (纸本)0769523803

This paper evaluates the performance of a novel Yew-Oriented Parallel Programming style for parallel programming on cluster computers. View-Oriented Parallel Programming is based on distributed shared memory which is friendly and easy for programmers to use. It requires the programmer to divide shared data into views according to the memory access pattern of the parallel algorithm. One of the advantages of this programming style is that it offers the performance potential for the underlying distributed shared memory system to optimize consistency maintenance. Also it allows the programmer to participate in performance optimization of a program through wise partitioning of the shared data into views. Experimental results demonstrate a significant performance gain of the programs based on the View-Oriented Parallel Programming style.

关键词： distributed shared memory view-based consistency view-oriented parallel programming cluster computing

来源：评论

学校读者我要写书评

暂无评论

Reconfigurable consistency algorithm

Reconfigurable consistency algorithm

引用

8th International Conference on High-Performance Computing in Asia-Pacific Region

作者： Pousa, Christiane V. Goes, Luis F. W. Martins, Carlos A. P. S. Pontificia Univ Catolica Minas Gerais Grad Program Elect Engn Computat & Digital Syst Grp Belo Horizonte MG Brazil

ISBN: (纸本)0769524869

In this paper, we propose, present and analyze the behavior and the performance of a reconfigurable algorithm for shared objects consistency management in distributed systems. Object sharing allows nodes to concurrently/parallel access a same set of replicated objects. However, it is necessary that the nodes know when and how to do these accesses, avoiding inconsistencies in the objects state. The RCA (Reconfigurable Consistency Algorithm) is a reconfigurable algorithm that guarantees the objects consistency. This algorithm modifies its behavior and structure according to the changes in the workload and distributed systems parameters. The paper shows that: the use of RCA generates flexibility and improves the performance in 30%, on average.

关键词： consistency algorithm objects distributed shared memory

来源：评论

学校读者我要写书评

暂无评论

A reconfigurable computing environment for urban traffic systems

A reconfigurable computing environment for urban traffic sys...

引用

19th European Conference on Modelling and Simulation (ECMS 2005)

作者： Khalil, M Peytchev, E Nottingham Trent Univ Sch Comp & Informat Nottingham NG1 4BU England

ISBN: (纸本)1842331159

This paper presents a reconfigurable computing environment for building hierarchical traffic telematics distributed systems based on non-locking distributed shared memory algorithm. The algorithm aims mainly at minimising the total amount of time for data retrieval in network of work-stations, considering the point of view of distributed traffic modules. The framework presented in this paper adopts a non-locking model to achieve the required performance. The presented framework develops further the successful features of DINE (developed and designed at SOCI, NTU) and at the same time avoids its shortcomings. The experimental results show that the new framework outperforms the old design of the system.

关键词： telematics distributed shared memory non-locking partial replication reconfigurable

来源：评论

学校读者我要写书评

暂无评论

A computing environment for urban traffic systems

A computing environment for urban traffic systems

引用

7th International Conference on Advanced Communication Technology

作者： Khalil, M Peytchev, E Al-Dabass, D Nottingham Trent Univ Sch Comp & Informat Nottingham NG1 4BU England

ISBN: (纸本)8955191235

This paper presents a computing environment for building hierarchical traffic telematics distributed systems based on non-locking distributed shared memory algorithm. The algorithm aims mainly at minimising the total amount of time for data retrieval in network of workstations, considering the point of view of distributed traffic modules. The framework presented in this paper adopts a non-locking model to achieve the required performance. The presented framework develops further the successful features of DIME (developed and designed at SOCI, NTU) and at the same time avoids its shortcomings. The experimental results show that the new framework outperforms the old design of the system.

关键词： telematics distributed shared memory nonlocking partial replication

来源：评论

学校读者我要写书评

暂无评论

memory sharing for interactive ray tracing on clusters

Memory sharing for interactive ray tracing on clusters

引用

Symposium on Parallel Graphics and Visualization

作者： DeMarle, DE Gribble, CP Boulos, S Parker, SG Univ Utah Sci Comp & Imaging Inst Salt Lake City UT 84112 USA

关键词： scientific visualization out-of-core rendering distributed shared memory ray tracing cache miss reduction

来源：评论

学校读者我要写书评

暂无评论

RCM - A multi-layered reconfigurable cluster middleware

RCM - A multi-layered reconfigurable cluster middleware

引用

14th Euromicro Conference on Parallel, distributed and Network-Bases Processing

作者： Nagel, R Rauber, T Univ Bayreuth Dept Comp Sci Bayreuth Germany

ISBN: (纸本)076952513X

DSM systems provide an easy-to-use programming model for parallel and distributed systems, but it is sometimes difficult to reach the performance characteristics of low-level message-passing programs, in particular if these have been optimized towards a specific architecture. In this article, we propose a multi-layered realization of a DSM system which provides different programming abstractions, including a level which allows an explicit control of the data placement. The programmer can select an appropriate level of abstraction for his application and it is even possible to mix program parts realized at different abstraction levels. The article gives a description of the multi-layered model, describes a prototype realization of the system and presents some preliminary experimental results on a heterogeneous system.

关键词： distributed shared memory shared virtual memory cluster computing task parallelism middle-ware

来源：评论

学校读者我要写书评

暂无评论

Binding time in distributed shared memories for generic patterns of memory references

引用

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS 2004年第8期E87D卷 2148-2151页

作者： Kong, J Lee, G Univ Illinois Chicago IL USA

Performance of three binding schemes for memory local to a node is evaluated. Since a large number of cache misses can occur in a large (relative to the cache size) working set, binding at a page fault time alone cannot efficiently utilize locality of reference at the local memory. In a small working set, the address bound to the local memory at a node miss time is not effective due to low cache miss rates. Our simulation shows that binding at a cache miss time achieves up to 3.1 times and 2.4 times performance of the schemes of binding at a page fault time and at a node miss time respectively.

关键词： cc-NUMA COMA distributed shared memory multiprocessor

来源：评论

学校读者我要写书评

暂无评论

Proteus: an efficient runtime reconfigurable distributed shared memory system

引用

JOURNAL OF SYSTEMS AND SOFTWARE 2001年第3期56卷 247-260页

作者： Ueng, JC Shieh, CK Liang, TY Chang, JB Natl Cheng Kung Univ Dept Elect Engn Tainan 701 Taiwan

This paper describes Proteus, a distributed shared memory (DSM) system which supports runtime node reconfiguration. Proteus allows users to change the node set during the execution of a DSM program. The capability of node addition allows users to further shorten the execution time of their DSM programs by dynamically adding newly available nodes to the system. Furthermore, competition for resources between system users and computer owners can be avoided by dynamically deleting nodes from the system. To make the system adapt to the node configuration efficiently, Proteus employs several techniques, including adaptive workload redistribution, affinity page movement, and forced update. Proteus supports both sequential consistency and release consistency. It provides an object-oriented parallel programming environment. This paper describes the design and implementation of node reconfiguration in Proteus, and presents the performance of the system. Experimental results indicate that Proteus can further improve the performance of the tested programs by taking advantage of node reconfiguration. Our results further demonstrate that the techniques employed in Proteus minimize communication and overhead. (C) 2001 Elsevier Science Inc. All rights reserved.

关键词： distributed shared memory workload redistribution page movement synchronization

来源：评论

学校读者我要写书评

暂无评论

Exploring virtual network selection algorithms in DSM cache coherence protocols

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 2004年第8期15卷 699-712页

作者： Chaudhuri, M Heinrich, M Cornell Univ Comp Syst Lab Ithaca NY 14853 USA Univ Cent Florida Sch Comp Sci Orlando FL 32816 USA

distributed shared memory (DSM) multiprocessors typically require disjoint networks for deadlock-free execution of cache coherence protocols. This is normally achieved by implementing virtual networks with the help of virtual channels or virtual lanes multiplexed on a single physical network. To keep the coherence protocol simple, messages are usually assigned to virtual lanes in a predefined static manner based on a cycle-free lane assignment dependence graph. However, this static split of virtual networks ( such as request and reply networks) may lead to underutilization of certain virtual networks while saturating the other networks. In this paper, we explore different static and dynamic schemes to select the virtual lanes for outgoing messages and mix the load among them without restricting any particular type of message to be carried only by a particular virtual network. We achieve this by exposing the selection algorithms to the coherence protocol itself, so that it can inject messages into selected virtual lanes based on some local information, and still enjoy deadlock-freedom. Our execution-driven simulation on five applications from the SPLASH-2 suite shows that as the system scales, the virtual network selection algorithms play an important role. For 128-node systems, our dynamic selection algorithm speeds up parallel execution by as much as 22 percent over an optimized baseline system running a modified SGI Origin 2000 protocol. We also explore how network latency, the number of message buffers per virtual lane, and the depth of network interface output queues affect the relative performance of various virtual lane selection algorithms.

关键词： distributed shared memory cache coherence protocol virtual network deadlock-freedom

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：