Given a graph G = (V, E), an independent set I is a subset of vertices, such that no two vertex in I share a common edge. A maximal independent set (MIS) is an independent set such that there is no way we can add any ...
详细信息
ISBN:
(纸本)9781450348393
Given a graph G = (V, E), an independent set I is a subset of vertices, such that no two vertex in I share a common edge. A maximal independent set (MIS) is an independent set such that there is no way we can add any other vertex such that the property of independent set remains intact. Different MIS heuristics are used in parallel applications for identifying the subset of independent tasks. Traditional MIS heuristics aim to generate minimum number of MIS i.e. number of parallel steps for the particular applications. However, if MIS produced have skewed sizes, hardware resource utilization becomes inefficient. parallelcomputing has become a key feature for improving performance in these days. A task is broken down into various subtasks and each sub task is executed in parallel on various processors. This is where generating and balancing the maximal independent sets come into play. In this paper, in the context of parallel application, we have proposed a heuristic for balancing the maximal independent sets using Hadoop after dividing the graph into various maximal independent sets. We have verified through experiment that our heuristic balances MIS' without much increasing the number of MIS generated.
The notion of verifiable database (VDB) plays an important role in secure outsourcing of data storage, which allows a client, typically a resource-limited one, to outsource a large-scale database to an untrusted serve...
详细信息
ISBN:
(纸本)9781538637906
The notion of verifiable database (VDB) plays an important role in secure outsourcing of data storage, which allows a client, typically a resource-limited one, to outsource a large-scale database to an untrusted server and make retrieval and update queries. For each query request, the server responds with a query result and a proof that is used to verify the result. Many VDB schemes have been proposed to provide the verification of outsourced databases, however, the existing VDB schemes are impracticable because of their inefficiency of initialization and query, unacceptable public key size and inability of tampered record localization scope. Therefore, in this paper, we present a new VDB scheme based on the Vector Commitments Tree (VCT) which is a new extension of the traditional Vector Commitments (VC). Our VDB scheme is superior to exiting schemes in at least two aspects: one is that the efficiency of initialization and query is greatly improved, and the other is that hierarchical scalable verification and tampered record localization are realized.
Modern supercomputer applications are demanding high-performance storage resources in addition to fast computing resources. However, these storage resources, especially parallel shared filesystems, have become the Ach...
详细信息
ISBN:
(纸本)9781728190747
Modern supercomputer applications are demanding high-performance storage resources in addition to fast computing resources. However, these storage resources, especially parallel shared filesystems, have become the Achilles' heel of many powerful supercomputers. Due to the lack of mechanism of IO resource provisioning on the file server side, a single user's IO intensive work running on a small number of nodes can overload the metadata server and result in global filesystem performance degradation and even unresponsiveness. To tackle this issue, we developed an innovative tool, Optimal Overloaded IO Protection System (OOOPS). This tool is designed to control the IO workload from applications side. Supercomputer administrators can easily assign the maximum number of function calls of open() and stat() allowed per second. OOOPS can automatically detect and throttle intensive IO workload to protect parallel shared filesystems. It also allows supercomputer administrators to dynamically adjust how much metadata throughput one job can utilize during the job runs without interruption.
This paper starts by giving a brief overview of Jini and highlights the features that can be effectively used for network computing. It describes how Jini attributes can be used in describing and searching for network...
详细信息
ISBN:
(纸本)076950759X
This paper starts by giving a brief overview of Jini and highlights the features that can be effectively used for network computing. It describes how Jini attributes can be used in describing and searching for network computing services. A system for high performance network computing is then described.
In Machine learning (ML) the model we use is increasingly important, and the model's parameters, the key point of the ML, are adjusted through iteratively processing a training dataset until convergence. Although ...
详细信息
ISBN:
(纸本)9781509042975
In Machine learning (ML) the model we use is increasingly important, and the model's parameters, the key point of the ML, are adjusted through iteratively processing a training dataset until convergence. Although data-parallel ML systems often engage a perfect error tolerance when synchronizing the model parameters for maximizing parallelism, the synchronization of model parameters may delay in completion, a problem that generally gets worse at a large scale. This paper presents a Bounded Asynchronous parallel (BAP) model of computation that allows computations using stale model parameters in order to reduce synchronization overheads. In the meanwhile, our BAP model ensures theoretical convergence guarantees for large scale data-parallel ML applications. This model permits distributed workers to use the stale parameters storing in the local cache, instead of waiting until the Parameter Server (PS) produces a new version. This expressively reduces the time workers spend on waiting. Furthermore, the BAP model guarantees the convergence of ML algorithm by bounding the maximum distance of the stale parameters. Experiments conducted on 4 cluster nodes with up to 32 GPUs showed that our model significantly improved the proportion of computing time relative to the waiting time and led to 1.2-2 X speedup. Besides, we elaborated how to choose the staleness threshold when considering the tradeoff between Efficiency and Speed.
Designing fast singular value decomposition (SVD) is significantly interesting in applications. The random direct SVD (RSVD) has provided a fast scheme to compute the well approximate SVD by unilateral randomized samp...
详细信息
ISBN:
(纸本)9781509050819
Designing fast singular value decomposition (SVD) is significantly interesting in applications. The random direct SVD (RSVD) has provided a fast scheme to compute the well approximate SVD by unilateral randomized sampling. In this paper, we present an efficient random algorithm in a bilateral sampling way. We also prove that the proposed algorithms can be bounded well and have less computational complexity compared to RSVD when the objective matrix is approximately square. Numerical experiments on graph Laplacian and Hilbert matrix demonstrate the efficiency and stability of the proposed methods.
Internet of Things is a new emerging technology that promises a new era of Internet through encompassing seamlessly physical and digital worlds in one single intelligent ecosystem. This goal is achieved by interconnec...
详细信息
ISBN:
(纸本)9781538637906
Internet of Things is a new emerging technology that promises a new era of Internet through encompassing seamlessly physical and digital worlds in one single intelligent ecosystem. This goal is achieved by interconnecting a large number of smart objects from the physical word such as smartphones, sensors, robots, connected cars, etc., to Internet. Nowadays, with the advent of Internet of Things, we need efficient mechanisms to remotely control IoT smart actuators by users and controllers using smartphones and IoT devices. This arises particularly in industrial Cyber-Physical Systems to supervise industrial processes. However, the complex environment of IoT systems makes this task very difficult to achieve regarding the number of connected objects and their resource limitation. In this paper, we tackle the problem of remote secure control of IoT actuators. We propose a distributed lightweight fine-grained access control based on Attribute Based Encryption mechanism and one way hash chain. We conducted security analysis and formal verification using AVISPA. The results demonstrated that our scheme is secure against various attacks. Moreover, the simulation results demonstrated the scalability and the efficiency of our solution, which saves substantially energy consumption and computation costs.
This paper describes the network communication occurring within a large-scale, interactive, distributed 3D environment, which supports closed coupled collaboration. Its main goals are the maintenance of low latency du...
详细信息
ISBN:
(纸本)1880843528
This paper describes the network communication occurring within a large-scale, interactive, distributed 3D environment, which supports closed coupled collaboration. Its main goals are the maintenance of low latency during user interaction and fast multicasting of messages in order to fulfill consistency requirements. This is achieved by sophisticated message distribution techniques peer-to-peer connections between interacting clients and a global hierarchical communication topology.
In recent years grid which facilitates the sharing and integration of large scale heterogenous resources has been recognized as the future framework of distributedcomputing. Modern software is plagued by security fla...
详细信息
ISBN:
(纸本)9783642194221
In recent years grid which facilitates the sharing and integration of large scale heterogenous resources has been recognized as the future framework of distributedcomputing. Modern software is plagued by security flaws at many levels. Implementation of an autonomic system provides an inherent self-managing capability to the system overcoming the shortcomings of the manual system. This paper presents a new self-protection model- SASM based upon few principles of genetics.
Escalating global power(energy) demands and the need to avail it in a reliable, efficient manner has led to the modernization of legacy and current power system grids into Smart grid (SGs) equivalents his article prop...
详细信息
暂无评论