the problem of inverting matrices is one that occurs in some problems of practical importance. this paper introduces and evaluates the block algorithm for high performance matrix inversion on the Cell Broadband Engine...
详细信息
ISBN:
(纸本)9781424446001
the problem of inverting matrices is one that occurs in some problems of practical importance. this paper introduces and evaluates the block algorithm for high performance matrix inversion on the Cell Broadband Engine (Cell/B.E.) processor. the Cell/B.E. is a heterogeneous multi-core processor on a single-chip jointly developed by Sony, Toshiba and IBM, which has a very high speed of the single precision floating-point arithmetic. the discussed matrix inversion algorithm is a combination of the block Algebraic Path Problem algorithm and the well-known block matrix inversion algorithm based on the LU decomposition. For relatively big matrices, this combined block algorithm spends the most time in computing matrix-matrix multiplication of blocks and achieves 149.4 Gflop/s on Cell/B.E., when PPE and six SPEs of PlayStation3 are used, or 93.4% of the aggregated double (PPE) and single (SPEs) precision peak performance, which is 160.0 Gflop/s.
the Grid is an heterogeneous and dynamic environment which enables distributed computation. this makes it a technology prone to failures. Some related work uses replication to overcome failures in a set of independent...
详细信息
ISBN:
(纸本)9781424446001
the Grid is an heterogeneous and dynamic environment which enables distributed computation. this makes it a technology prone to failures. Some related work uses replication to overcome failures in a set of independent tasks, and in workflow applications, but they do not consider possible resource limitations when scheduling the replicas. In this paper, we focus on the use of task replication techniques for workflow applications, trying to achieve not only tolerance to the possible failures in an execution, but also to speed up the computation without demanding the user to implement an application-level checkpoint, which may be a difficult task depending on the application. Moreover, we also study what to do when there are not enough resources for replicating all running tasks. We establish different priorities of replication depending on the graph of the workflow application, giving more priority to tasks with a higher output degree. We have implemented our proposed policy in the GRID superscalar system, and we have run the fastDNAml as an experiment to prove our objectives are reached. Finally, we have identified and studied a problem which may arise due to the use of replication in workflow applications: the replication wait time.
We consider snap-stabilization in message-passing systems. Snap-stabilization permits to design protocols that withstand transient faults: Any computation that is started after faults cease immediately satisfies the e...
详细信息
ISBN:
(纸本)9783540922940
We consider snap-stabilization in message-passing systems. Snap-stabilization permits to design protocols that withstand transient faults: Any computation that is started after faults cease immediately satisfies the expected specification. Our contribution is twofold, as we demonstrate that in message passing systems (i) snap-stabilization is impossible for nontrivial problems if we consider channels with finite yet unbounded capacity, and (ii) snap-stabilization becomes possible in the same setting with bounded-capacity channels. the latter contribution is constructive, as we propose two snap-stabilizing protocols.
A key prerequisite for residential energy conservation is knowing when and where energy is being spent. Unfortunately, the current generation of energy reporting devices only provide partial and coarse grained informa...
详细信息
ISBN:
(纸本)9781605584317
A key prerequisite for residential energy conservation is knowing when and where energy is being spent. Unfortunately, the current generation of energy reporting devices only provide partial and coarse grained information or require expensive professional installation. this limitation stems from the presumption that calculating per-appliance consumption requires per-appliance current measurements. However, since appliances typically emit measurable signals when they are consuming energy, we can estimate their consumption using indirect sensing. this paper presents ViridiScope, a fine-grained power monitoring system that furnishes users with an economical, self-calibrating tool that provides power consumption of virtually every appliance in the home. ViridiScope uses ambient signals from inexpensive sensors placed near appliances to estimate power consumption, thus no inline sensor is necessary. We use a model-based machine learning algorithm that automates the sensor calibration process. through experiments in a real house, we show that ViridiScope can estimate the end-point power consumption within 10% error.
Just recently emerged a new paradigm for internet-based software systems which is called as cloud computing. the cloud provides scalable processing power and several kinds of connectable services. this distributed arc...
详细信息
In this paper, closed-loop quasi-orthogonal space time block coding (QO-STBC) is exploited within a four relay node transmission scheme to achieve full-rate and increase the available diversity gain provided by earlie...
详细信息
ISBN:
(纸本)9781424453337
In this paper, closed-loop quasi-orthogonal space time block coding (QO-STBC) is exploited within a four relay node transmission scheme to achieve full-rate and increase the available diversity gain provided by earlier two relay approaches. the problem of imperfect synchronization between relay nodes is overcome by applying a parallel interference cancellation (PIC) detection scheme at the destination node. Bit error rate simulations confirm the advantages of the proposed methodology for a range of levels of imperfect synchronization and that only a small number of iterations is necessary within the PIC detection.
In Telecom Social Network, the Context-based Prediction Analysis is an important topic, and computing of some context parameters using great amount of data is a problem so far. Architecture of DMG ( Data Mining Grid i...
详细信息
ISBN:
(纸本)9781424448982
In Telecom Social Network, the Context-based Prediction Analysis is an important topic, and computing of some context parameters using great amount of data is a problem so far. Architecture of DMG ( Data Mining Grid is proposed and prototype of MDG is designed to solve the computing problem. Two of the important issues in DMG, which are the design of the workflow service in DMG and the distributed data mining algorithm, are investigated. Finally a sample application in telecom field Customer Chum Context Prediction Analysis is investigated and illustrated. In the sample application, Centrals, which is computed by parallel algorithm on DMG, is an important measure in Telecom Social Network. And it is an important context parameter proved by the experiment, which benefits to enhance the precise of context-based prediction of the customer chum in telecom social network.
computing of some parameters using great amount of data is a great challenge in data mining fields so far. DMG (Data Mining Grid) is designed to solve the computing problem. the design of the workflow service in DMG a...
详细信息
ISBN:
(纸本)9781424448982
computing of some parameters using great amount of data is a great challenge in data mining fields so far. DMG (Data Mining Grid) is designed to solve the computing problem. the design of the workflow service in DMG and the distributed data mining algorithm, are investigated. A sample application in telecom field Customer Value Analysis is and illustrated. In the sample application, Clique, which is Computed by parallel algorithm on data mining grid system, is an important measure in Telecom Social Network, which benefits to enhance the precise of context-based prediction of the customer value in telecom social network and work out information in expected duration. Telecom operators can take measures to gain more profit.
Over the past decade, scheduling in distributedcomputing system has been an active research. However, it is still difficult to find an optimal scheduling algorithm to achieve load balancing for a specific scientific ...
详细信息
ISBN:
(纸本)9781424437719
Over the past decade, scheduling in distributedcomputing system has been an active research. However, it is still difficult to find an optimal scheduling algorithm to achieve load balancing for a specific scientific application which is executed in an unpredictable environment. this is due to the complex nature of the application which changes during runtime and due to the dynamic nature and unpredictability of the computational environment. this paper addresses these issues by presenting a Simulated Annealing (SA) approach as an optimizer which is an improved version of EG-EDF with Tabu Search optimizer. Instead of using Tabu Search, this work used SA to optimize the scheduling algorithm. the scheduling algorithms have been evaluated using three main criteria;number of delayed jobs, makespan time and total tardiness. Our results show the improvements to the main criteria mentioned.
Digital homes provide the best services to home's occupants by using modern devices, home appliances that integrate transparently and simplify usability in the home. However, the realization in the real environmen...
详细信息
ISBN:
(纸本)9788955191387
Digital homes provide the best services to home's occupants by using modern devices, home appliances that integrate transparently and simplify usability in the home. However, the realization in the real environment is difficult and takes a long time from building the scratch. thus, to support the implementation in the real smart home, it is necessary to demonstrate that thing can be done in the simulator which deals with virtual appliances and devices models the real smart home environment. In this paper, we propose ISS, an Interactive Smart home Simulator system focusing on controlling and simulating the behavior of an intelligent house. the developed system aims to provide architects, designers a simulation and useful tool for understanding. the interaction between environment, people and the impact of embedded and pervasive technology on in daily life. In this research, the smart house is considered as an environment made up of independent and distributed devices interacting to support user's goals and tasks. therefore, by using ISS, the developer can realize the relationship among virtual home space, surrounded environment, users and home appliances.
暂无评论