This paper describes a novel approach to deterministic multithreading for active replication of Java objects. Unlike other existing approaches, the presented deterministic thread scheduler fully supports the native Ja...
详细信息
ISBN:
(纸本)9780889866386
This paper describes a novel approach to deterministic multithreading for active replication of Java objects. Unlike other existing approaches, the presented deterministic thread scheduler fully supports the native Java synchronisation mechanisms, including reentrant locks, condition variables, and time bounds on wait operations. Furthermore, this paper proposes source-code transformation as a novel approach for intercepting Java synchronisation statements. This allows the reuse of existing object implementations and simplifies application development.
Emerging multiple display infrastructures provide users with a large number of semi-public and private displays. Selecting what information to present on which display here becomes a real issue, especially when multip...
详细信息
ISBN:
(纸本)9780889866386
Emerging multiple display infrastructures provide users with a large number of semi-public and private displays. Selecting what information to present on which display here becomes a real issue, especially when multiple users with diverging interests have to be considered. This especially holds for dynamic ensembles of displays. We propose to cast the Display Mapping problem as an optimization task. We develop an explicit criterion for the global quality of a display mapping and then describe a distributed algorithm based on the GRASP framework that is able to approximate the global optimum through local interaction between display devices. We claim that such a distributed optimization approach, based on the definition of an explicit global quality measure, is a general concept for achieving coherent ensemble behavior.
For search-intensive applications such as data mining and bioinformatics, a SIMD Processor Array on a Chip may be an effective architecture, and if the application is control-intensive, a Multiple SIMD (MSIMD) archite...
详细信息
ISBN:
(纸本)9780889866386
For search-intensive applications such as data mining and bioinformatics, a SIMD Processor Array on a Chip may be an effective architecture, and if the application is control-intensive, a Multiple SIMD (MSIMD) architecture may further increase processor utilization. In this paper, we describe the implementation of an associative MSIMD architecture on the MASC Processor. The MASC Processor implemented using FPGAs, is easily scalable, and dynamically assigns tasks to Processing Elements as the program executes.
Volunteer computing is an innovative approach to high performance computing that relies on volunteers who donate their personal computers' unused resources to a computationally intensive research project. Prominen...
详细信息
ISBN:
(纸本)9780889866386
Volunteer computing is an innovative approach to high performance computing that relies on volunteers who donate their personal computers' unused resources to a computationally intensive research project. Prominent volunteer computing projects include SETI@home, Folding@Home, and The Great Internet Mersenne Prime Search (GIMPS). Many volunteer computing projects are built upon a volunteer computing framework that abstracts functionality that is common to all volunteer computing projects, such as network communications, database access, and project management. These volunteer computing frameworks tend to be complex, limiting, and difficult to use. We have designed and implemented a new volunteer computing framework called the Simple Light-weight Infrastructure for Network computing (SLINC) that addresses the disadvantages we identified with existing frameworks. SLINC is a flexible and extensible volunteer computing framework that will enable researchers to more easily build volunteer computing projects.
This paper research on how to select a subtree with exactly k leaves and a diameter of at most 1, which minimizes the distance from the farthest vertex to the subtree. We call such a subtree (k, l)-center of a tree ne...
详细信息
ISBN:
(纸本)9780889866386
This paper research on how to select a subtree with exactly k leaves and a diameter of at most 1, which minimizes the distance from the farthest vertex to the subtree. We call such a subtree (k, l)-center of a tree network. In this paper, an efficient parallel algorithm is proposed for finding a (k, l)-center of a tree network. This algorithm performs on the EREW PRAM in O(log n) time using O(n) work.
distributed service architectures are mandatory to handle the platform scale and dynamicity hindering the development of grid and P2P applications. These large-scaled distributed applications are difficult to design, ...
详细信息
ISBN:
(纸本)9780889866386
distributed service architectures are mandatory to handle the platform scale and dynamicity hindering the development of grid and P2P applications. These large-scaled distributed applications are difficult to design, develop and tune because of both theoretical and practical issues. This paper presents the GRAS framework that allows developers to first implement and experiment with such an infrastructure in simulation, benefiting from a controlled environment. The infrastructure can then be deployed in situ without code modification. We detail our design goals, and contrast them with the state of the art. We study the exchange of a message (from the Pastry protocol) using either GRAS 0 or several other solutions. We quantify both the code complexity and the performance and find that GRAS performs better according to both metrics.
The efficient characterization of adaptive parallel applications is usually challenging due to their complexity and large problem size. Unlike traditional profiling approaches which target the tracing of events or det...
详细信息
ISBN:
(纸本)9780889866386
The efficient characterization of adaptive parallel applications is usually challenging due to their complexity and large problem size. Unlike traditional profiling approaches which target the tracing of events or determining performance parameters for subroutines, the approach described in this paper attempts to discover the inherent adaptivity of parallel applications mapped to the computation domain/mesh, which are independent of runtime environment, so as to aid in the performance tuning of parallel applications, especially dynamic load balancing and repartitioning. Our profiling scheme only requires one-time execution of the target program on any platform to generate a sequence of traces with timestamps. The traces can then be fed to simulations under various system configurations while independent of the real application. Preliminary experiments have been performed to evaluate the proposed profiling techniques.
As new genes are sequenced, it is common for molecular biologists to compare the new gene's DNA to known sequences. One simple form of DNA sequence comparison is done by solving the Longest Common Subsequence (LCS...
详细信息
ISBN:
(纸本)9780889866386
As new genes are sequenced, it is common for molecular biologists to compare the new gene's DNA to known sequences. One simple form of DNA sequence comparison is done by solving the Longest Common Subsequence (LCS) problem. In this paper, we propose a parallel algorithm and specialized FPGA-based processor (the associative ASC Processor with reconfigurable 2D mesh) to solve the exact and approximate match LCS problems. This solution uses inexpensive hardware and can be reconfigured as new analysis techniques are developed, making it particularly attractive for processing biosequences.
The Replica Placement Problem (RPP) aims at selecting the nodes for duplicating data objects in order to optimize their access. Even though a lot of work already exists on RPP, the issue of implementing the resulting ...
详细信息
ISBN:
(纸本)9780889866386
The Replica Placement Problem (RPP) aims at selecting the nodes for duplicating data objects in order to optimize their access. Even though a lot of work already exists on RPP, the issue of implementing the resulting allocation scheme is typically overlooked. In this paper we introduce the Replica Transfer Scheduling Problem (RTSP), briefly stated as: given a network of servers with limited storage capacitv, a set of data objects and two replication schemes x(old) and X-new, find a schedule of object transfers and deletions for implementing X-old based on X-old with minimum communication cost. Given that this problem is NP-complete, we introduce several different heuristics to solve it, and evaluate them via simulations.
Dynamic server resource allocation to services on networks, or utility computing, is a powerful technology to provide required computer resources for multiple service providers at low cost. Virtual machine (VM) techno...
详细信息
ISBN:
(纸本)9780889866386
Dynamic server resource allocation to services on networks, or utility computing, is a powerful technology to provide required computer resources for multiple service providers at low cost. Virtual machine (VM) technology can be combined with utility computing to further improve server resource utilization. An important technical issue about VM-based utility computing is optimal placement of VMs on physical server nodes, because performance of services may be seriously affected by VM placement. We address this issue by introducing an on-line placement algorithm on the basis of performance influence among services. We have implemented this algorithm and evaluated on a simulated environment. The results have shown that the proposed mechanism can get roughly 25% better score calculated by the placement rules compared with a random placement algorithm.
暂无评论