computing reversal distance of two signed permutations has gained increasing attention over the last decade withthe study of genome rearrangements in computational molecular biology. In this paper, we present a paral...
详细信息
Graph/hypergraph partitioning models and methods have been successfully used to minimize the communication among processors in several parallelcomputingapplications. parallel sparse matrix-vector multiplication (SpM...
详细信息
ISBN:
(数字)9783642551956
ISBN:
(纸本)9783642551956
Graph/hypergraph partitioning models and methods have been successfully used to minimize the communication among processors in several parallelcomputingapplications. parallel sparse matrix-vector multiplication (SpMxV) is one of the representative applicationsthat renders these models and methods indispensable in many scientific computing contexts. We investigate the interplay of the partitioning metrics and execution times of SpMxV implementations in three libraries: Trilinos, PETSc, and an in-house one. We carry out experiments with up to 512 processors and investigate the results with regression analysis. Our experiments show that the partitioning metrics influence the performance greatly in a distributed memory setting. the regression analyses demonstrate which metric is the most influential for the execution time of the libraries.
Leadership-scale scientific simulations running as tens of thousands of tightly-coupled MPI processes are vulnerable to interruption due to a single process or node failure. Due to the dependence of each state calcula...
详细信息
Withthe development of information technology, real-time data stream processing(RTDSP) has become a popular research topic. the first step of RTDSP is collecting data, requiring a data collector to receive data from ...
详细信息
ISBN:
(纸本)9781538637906
Withthe development of information technology, real-time data stream processing(RTDSP) has become a popular research topic. the first step of RTDSP is collecting data, requiring a data collector to receive data from the source and send them to the sink. Apache Flume, a distributed and reliable framework, used for this purpose, has some limitations and drawbacks on load balancing and storage. In this paper, we aim to improve performance and availability for collecting unstable real-time big data stream. So we propose a new load balancing strategy based on the free memory size and a storage strategy of integration memory channel withthe multi-file channel to reduce the overhead of disk and network. Finally, the experimental results show that the availability and performance are improved under the condition of a poor network, high availability requirements, intense competition in memory resources and large data size. Specifically, the availability is higher than 99.999%, and the performance can be improved by 10%-50% under different conditions.
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
When datasets are distributed on different sources, finding out their intersection while preserving the privacy of the datasets is a widely required task. In this paper we address the Privacy Preserving Set Intersecti...
详细信息
ISBN:
(纸本)9780769530499
When datasets are distributed on different sources, finding out their intersection while preserving the privacy of the datasets is a widely required task. In this paper we address the Privacy Preserving Set Intersection (PPSI) problem, in which each of the N parties learns no elements other than the intersection of their N private datasets. We propose an efficient protocol in the malicious model, where the adversary may control arbitrary number of parties and execute the protocol for its own benefit. A related work in [12] has a correctness probability of (N-1/N)(N) (N is the size of the encryption scheme's plaintext space), a computation complexity of O(N(2)S(2)lgN) (S is the size of each party's data set). Our PPSI protocol in the malicious model has a correctness probability of (N-1/N)(N-1), and achieves a computation cost of O(c(2)S(2)lgN) (c is the number of malicious parties and c <= N-1).
Large scale loosely coupled PCs can organize clusters and form desktop computing grids on sharing each processing power;power of PCs, transaction distributions and load balancing characterize the performance of the co...
详细信息
ISBN:
(纸本)9783642131356
Large scale loosely coupled PCs can organize clusters and form desktop computing grids on sharing each processing power;power of PCs, transaction distributions and load balancing characterize the performance of the computing grids. this paper describes the design methodologies of workload management in distributed desktop computing grids. Based on the prototype experiment, several simulations were performed;several centralized and decentralized algorithms for location policy were examined, and the design methodologies for distributed desktop computing grids are derived from the simulation results. the methodologies for domains, language and control algorithms for computing grids are described. the language for distributed desktop computing is designed to accomplish the design methodologies.
Non-periodic bursts are prevalent in workloads of large scale applications. Existing workload models do not predict such non-periodic bursts very well because they mainly focus on repeatable base functions. We begin b...
详细信息
the management of huge amounts of data distributed across multiple sites has become a necessity more and more demanding. Peer-to-peer systems (P2P) can afford the requirements of managing, indexing, searching and anal...
详细信息
ISBN:
(纸本)9781467394734
the management of huge amounts of data distributed across multiple sites has become a necessity more and more demanding. Peer-to-peer systems (P2P) can afford the requirements of managing, indexing, searching and analyzing data with scalability and self-organization. Until now, most efforts have focused primarily on improving the number of hops and structure maintenance messages. However, the non-uniform distribution of data and the hierarchical structures, together with heavy load, can cause unbalanced traffic load. In this paper we improve our previous work on the overlay structure G-Grid, merging it with a Small World network. the Small World networks make a compromise between order and randomness, they are derived from social networks and show an almost uniform traffic distribution. Experiments show how this new hybrid structure obtains the best performance in traffic distribution.
A new mathematical morphology-based algorithm is proposed to achieve automatic center location of non-eye typhoon. the center of a non-eye typhoon is near the geometric center of the cloud system and has higher temper...
详细信息
ISBN:
(纸本)0780378407
A new mathematical morphology-based algorithm is proposed to achieve automatic center location of non-eye typhoon. the center of a non-eye typhoon is near the geometric center of the cloud system and has higher temperature. For each infrared satellite cloud image, the locating procedures are as follows: a) noises filtering, b) main cloud systems segmenting, c) center locating and d) multispectral image verification. the algorithms are based on operations of mathematical morphology, and designed in IBM SP2 massively parallel computer. the experiment results show that the algorithm locates the centers of most non-eye typhoons successfully and achieves faster, more precise and non-human interactive non-eye typhoon center positioning.
暂无评论