this paper addresses the optimization of parallel simulators for large-scale parallel systems and applications. Such simulators are often based on parallel discrete event simulation with conservative or optimistic pro...
详细信息
this paper addresses the optimization of parallel simulators for large-scale parallel systems and applications. Such simulators are often based on parallel discrete event simulation with conservative or optimistic protocols to synchronize the simulating processes. the paper considers how available future information about events and application behaviors can be efficiently extracted and further exploited to improve the performance of adaptive optimistic protocols. First, we extract information about future events and their dependencies in application traces to guide adaptive adjustments of time window in trace-driven parallel simulation. Second, we use information about application behaviors, specifically the iterative behavior found in many applications, to avoid the unnecessary adjustments of time window. these techniques are implemented in the BigSim simulator and tested by real-world and standard benchmark applications including Jacobi3D and HPL. the results show that our optimization approaches can reduce the execution times of simulation ranging from 11% up to 32%. Moreover, our methods are easy to implement and don't need to augment compilers or even modify the core codes of parallel simulators.
Secure distributed storage systems can achieve data security and fault resilience by providing encrypted and redundant data segments across multiple computing nodes. However, such systems might face severe performance...
详细信息
Secure distributed storage systems can achieve data security and fault resilience by providing encrypted and redundant data segments across multiple computing nodes. However, such systems might face severe performance bottlenecks when frequently tacking data with high consistency demand. this paper proposed a secure distributed storage system based on revised secret sharing scheme with internal padding. Data are saved in multiple coefficients of associated polynomials for high capacity. Shares generated from these polynomials are distributed and can be recollected to retrieve the original data. Post-generation data operations including data insertion, deletion, updating and appending do not require the complete share regeneration. Only affected data portions will be reprocessed and parallel processing is supported if they are interleaved. Performance analyses and experimental results have demonstrated the effectiveness and efficiency of the post-generation data operations.
One among the most influential and popular data mining methods is the k-Means algorithm for cluster analysis. Techniques for improving the efficiency of k-Means have been largely explored in two main directions. the a...
详细信息
One among the most influential and popular data mining methods is the k-Means algorithm for cluster analysis. Techniques for improving the efficiency of k-Means have been largely explored in two main directions. the amount of computation can be significantly reduced by adopting geometrical constraints and an efficient data structure, notably a multidimensional binary search tree (KD-Tree). these techniques allow to reduce the number of distance computations the algorithm performs at each iteration. A second direction is parallel processing, where data and computation loads are distributed over many processing nodes. However, little work has been done to provide a parallel formulation of the efficient sequential techniques based on KD-Trees. Such approaches are expected to have an irregular distribution of computation load and can suffer from load imbalance. this issue has so far limited the adoption of these efficient k-Means variants in parallelcomputing environments. In this work, we provide a parallel formulation of the KD-Tree based k-Means algorithm for distributed memory systems and address its load balancing issue. three solutions have been developed and tested. Two approaches are based on a static partitioning of the data set and a third solution incorporates a dynamic load balancing policy.
High level context recognition and situation detection are enabling technologies for unobtrusive mobile computing systems. Significant progress has been made in processing and managing context information, leading to ...
详细信息
High level context recognition and situation detection are enabling technologies for unobtrusive mobile computing systems. Significant progress has been made in processing and managing context information, leading to sophisticated frameworks, middlewares, and algorithms. Despite great improvements, context aware systems still require a significantly increased recognition accuracy for high-level context information on uncertain sensor data to enable the robust execution of context-aware applications. Recently Adaptable Pervasive Workflows (APF)s have been presented as innovative programming paradigm for mobile context-aware applications. We propose a novel Flow Context System (FlowCon) that builds upon APFs. FlowCon uses structural information from the APF to increase accuracy of uncertain high-level context information up to 49%. this way we make an important step to enable robust execution of mobile context-aware applications.
In distributed virtual environments (DVEs), maintaining a consistent view of the virtual world among all users is a primary task. Due to the resource limitations such as network capacity and computational power, the c...
详细信息
In distributed virtual environments (DVEs), maintaining a consistent view of the virtual world among all users is a primary task. Due to the resource limitations such as network capacity and computational power, the consistency of the virtual world cannot be guaranteed sometimes. In this paper, we try to address this problem in client-server-based DVEs by considering server side network capacity limitation. We look at the position update packets from client to server and show the impact of packet loss due to server side network capacity limitation on the consistency of the virtual world. We assume clients use Dead Reckoning (DR) models to control position updates. In order to adapt client updates withthe server side network capacity to avoid update packet loss, two server assistant DR threshold tuning policies for different updating schemas are proposed in this paper. the proposed tuning polices are validated to have good performance by simulation results.
Data-intensive parallelapplications on clouds need to deploy large data sets from the cloud's storage facility to all compute nodes as fast as possible. Many multicast algorithms have been proposed for clusters a...
详细信息
ISBN:
(纸本)9780769540399
Data-intensive parallelapplications on clouds need to deploy large data sets from the cloud's storage facility to all compute nodes as fast as possible. Many multicast algorithms have been proposed for clusters and grid environments. the most common approach is to construct one or more spanning trees based on the network topology and network monitoring data in order to maximize available bandwidth and avoid bottleneck links. However, delivering optimal performance becomes difficult once the available bandwidth changes dynamically. In this paper, we focus on Amazon EC2/S3 (the most commonly used cloud platform today) and propose two high performance multicast algorithms. these algorithms make it possible to efficiently transfer large amounts of data stored in Amazon S3 to multiple Amazon EC2 nodes. the three salient features of our algorithms are (1) to construct an overlay network on clouds without network topology information, (2) to optimize the total throughput dynamically, and (3) to increase the download throughput by letting nodes cooperate with each other. the two algorithms differ in the way nodes cooperate: the first `non-steal' algorithm lets each node download an equal share of all data, while the second `steal' algorithm uses work stealing to counter the effect of heterogeneous download bandwidth. As a result, all nodes can download files from S3 quickly, even when the network performance changes while the algorithm is running. We evaluate our algorithms on EC2/S3, and show that they are scalable and consistently achieve high throughput. Both algorithms perform much better than having each node downloading all data directly from S3.
An autoimmune disorder is a condition that occurs when the immune system mistakenly attacks and destroys healthy body parts. the epidemiology of these diseases is a matter of study and discussion, with many published ...
详细信息
An autoimmune disorder is a condition that occurs when the immune system mistakenly attacks and destroys healthy body parts. the epidemiology of these diseases is a matter of study and discussion, with many published results worldwide. We have critically collected more than 100 publications, analyzing autoimmune diseases' frequencies in various populations and ethnic groups. Simultaneously, we developed a web application in order to host the collected data. By utilizing Microsoft Silverlight technology, we have built a multimedia web frontend, in order to support a high level visualization of the data and the mining process. In parallel we continue the data mining process and the update of our database.
the proceedings contain 75 papers. the topics discussed include: a layered virtual organization architecture for grid;scalable contract net based resource allocation strategies for grids;an experimental analysis for m...
ISBN:
(纸本)9780769534435
the proceedings contain 75 papers. the topics discussed include: a layered virtual organization architecture for grid;scalable contract net based resource allocation strategies for grids;an experimental analysis for memory usage of GOS core;a data-parallel algorithm to reliably solve systems of nonlinear equations;an effective structure for algorithmic design and a parallel prefix algorithm on metacube;a parallel algorithm for block tridiagonal systems;parallelization and acceleration scheme of multilevel fast multipole method;parallel approximate multi-pattern matching on heterogeneous cluster systems;switch-based packing technique for improving token coherence scalability;location consistency model revisited: problem, solution and prospects;an enhancer of memory and network for cluster and its applications;honeycomb: a community-based system for distributed data integration and sharing;and set-to-set disjoint paths routing in dual-cubes.
A parallel version of a new automatic Harris-based corner detector is presented. A scheduler to dynamically and homogeneously distribute high computational workload on heterogeneous parallel architectures such as Grid...
详细信息
ISBN:
(纸本)9781424452910
A parallel version of a new automatic Harris-based corner detector is presented. A scheduler to dynamically and homogeneously distribute high computational workload on heterogeneous parallel architectures such as Grid systems has been implemented to speedup the whole procedure. Experimental results show the robustness of the underlying scheduler, which can be easily exploited in various automatic image analysis systems.
Finding a vast array of applications, the problem of computingthe convex hull of a set of sorted points in the plane is one of the fundamental tasks in pattern recognition, morphology and image processing. the main c...
详细信息
ISBN:
(纸本)9781424452910
Finding a vast array of applications, the problem of computingthe convex hull of a set of sorted points in the plane is one of the fundamental tasks in pattern recognition, morphology and image processing. the main contribution of this paper is to show a simple parallel algorithm for computingthe convex hull of a set of n sorted points in the plane and evaluate the performance on the dual quad-core processors. the experimental results show that, our implementation achieves a speed-up factor of approximately 7 using 8 processors. Since the speed-up factor of more than 8 is not possible, our parallel implementation for computingthe convex hull is close to optimal. Also, for 2 or 4 processors, we achieved a super linear speed up.
暂无评论