During the development and parametrization of 2D image-processing algorithms for surface inspection uses, you need to test a huge amount of image-data for each modification of the algorithms or parameters. For algorit...
详细信息
During the development and parametrization of 2D image-processing algorithms for surface inspection uses, you need to test a huge amount of image-data for each modification of the algorithms or parameters. For algorithm runtimes up to several seconds, this will take a long time. To speed up this process it is recommended to distribute the computation in a parallel computation environment. Compute grids, which use the unused resources of existing hardware are the most cost efficient way to solve this problem. the most existing grid-Concepts are based on flat connection structures with a scheduler on the top; for high job-rates the scheduler becomes the bottleneck of the whole system. Concepts to solve this problem organize the nodes in tree-structures to discharge the central scheduler. In heterogeneous Desktop-grids where the different nodes are widely distributedthe usually used random arrangement of the nodes in the tree-structure can be counterproductive, because the bandwidthes and latencies in a grid can be varying. In this paper we will show a solution to arrange the nodes of the grid optimized by bandwidth and latency, using modified spanning-tree algorithms, so that the average response time is reduced and in result of this the job-throughput of the Compute-grid is increased.
Intelligent Decision Analysis Platform is an advanced application of decision and analysis based on the construction of "State grid 186" in China. the platform is committed to achieve the goal of all kinds o...
详细信息
ISBN:
(纸本)9781479986477
Intelligent Decision Analysis Platform is an advanced application of decision and analysis based on the construction of "State grid 186" in China. the platform is committed to achieve the goal of all kinds of business managements and developments. It mainly focuses on providing scientific decisions by data mining and predicting after status data being analyzed. Withthe auxiliary support of the platform, power enterprises are capable of making better strategies and increasing revenue eventually. According to the characteristics of smart grid monitoring and the requirements of management efficiently as well as the demand of reliable storage of massive grid data, this paper puts forward the decision and analysis platform based on the parallel processing framework which is called Spark and the fault-tolerant abstraction for in-memory cluster computing embedded in Spark as it is known as Resilient distributed Datasets (RDDs). this paper gives detailed analysis of the feasibility and the advantages of the new method, as well as some unsolved problems.
grids can be considered as dominant platforms for large-scale parallel/distributedcomputing in science and engineering. Clouds allow users to acquire and release resources on-demand. Next generation computing environ...
详细信息
ISBN:
(纸本)9783901882449
grids can be considered as dominant platforms for large-scale parallel/distributedcomputing in science and engineering. Clouds allow users to acquire and release resources on-demand. Next generation computing environments will benefit from the combination of grid and Cloud paradigms providing frameworks that integrate traditional grid services with on-demand Cloud services. Nowadays, workflows are the preferred means for the combination of services into added value service chains representing functional business processes or complex scientific experiments. A promising way to manage effectively services composition in a dynamic and heterogenous environment is to make the workflow management framework able to self-adapt at runtime to changes in its environment and provide an uniform resource access mechanism over grid and Cloud infrastructures. Autonomic workflow management systems can support the runtime modification of workflows withthe aim of improving their performance and recover from faults determining and provisioning the appropriate mix of grid/Cloud services with requested QoS. this paper describes Sunflower an innovative P2P agent-based framework for configuring, enacting, managing and adapting workflows on hybrid grid-Cloud infrastructures. To orchestrate grid and Cloud services, Sunflower uses a bio-inspired autonomic choreography model and integrates the scheduling algorithm with a provisioning component that can dynamically launch virtual machines in a Cloud infrastructure to provide on-demand services in peak-load situations.
this project intends to use the recently developed OpenMP parallelcomputing method to integrate the minimum-maximum reclassification method withthe factor table path tree method. then the different types of images a...
详细信息
ISBN:
(数字)9798350350760
ISBN:
(纸本)9798350350777
this project intends to use the recently developed OpenMP parallelcomputing method to integrate the minimum-maximum reclassification method withthe factor table path tree method. then the different types of images are segmented. this method can decouple LU algorithm from other transient stability algorithms to realize parallel processing of multiple independent cores. Each core has high bandwidth and low latency reliable communication cut connections very close, capable of fine “multiple small” parallelism. According to the characteristics of hierarchical division of power grid, the communication time required in the implementation of this method is established, and the corresponding communication algorithm is given. this method can reduce the communication times required to solve the quadrative variables, improve the convergence of numerical simulation, and greatly improve the distributedparallelcomputing efficiency of transient stability. the experimental results show that the communication method proposed in this paper is feasible.
Nowadays, more researchers in the simulation field are interested in the use of a service-oriented architecture and component-based system in order to design more efficient large-scale distributed simulation systems. ...
详细信息
Nowadays, more researchers in the simulation field are interested in the use of a service-oriented architecture and component-based system in order to design more efficient large-scale distributed simulation systems. A Service-Oriented Architecture (SOA) is a collection of independent services with well-defined interfaces that can be accessed without knowledge of their implementation. A component-based approach separates the whole system into different parts based on their functionalities. In this paper, a framework for component-based distributed simulation on the grid (SOAr-DSgrid) is introduced. One of the major issues in designing the framework is to have efficient data FLow and control flow in the framework. Data flow describes the way that data moves in the system while control flow determines the order of activities that happen in the system. the rule engine is a critical part of the framework and is responsible for managing the distributed data flow and control FLow. the design and implementation of the rule engine is the major focus of this paper. the rule engine enables the framework to be more scalable as the components within the framework now communicate in a fully distributed manner.
Transaction scheduling and concurrency control are two important problems in a distributed environment. A lot of research has been focused on the problems recently, and many models and measures have been proposed. How...
详细信息
Transaction scheduling and concurrency control are two important problems in a distributed environment. A lot of research has been focused on the problems recently, and many models and measures have been proposed. However, some of them didn't fit for grid environment, as grid architecture is a new and distinguished from conventional distributedcomputing by its heterogeneous, dynamic and autonomous properties. In this paper, we propose a three-level transaction scheduling model for grid database, which not only shows the autonomy of grid but also lightens the pressure of communication and grid scheduler. We consider communication time of a transaction, user's priority and the number of DB sites accessed by the transaction as a priority of the transaction, and give scheduling algorithms for concurrency transaction, which not only supports correctness of the data, but also improves the effectiveness of the system. Petri nets are promising tools for modeling and analysis information processing systems that are characterized as being concurrent, parallel and distributed. We use a dynamic time colored Petri net (DTCPN) to model each level's transaction scheduling and get a DTCPN model for three-level transaction scheduling model of grid database using sharing composition. Finally, we get some important results such as miss rate of transactions by analyzing reachability of Petri nets.
Withthe immense computing power at our disposal, the numerical solution of partial differential equations (PDEs) is becoming a day-to-day task for modern computational scientists. However, the complexity of real-life...
详细信息
ISBN:
(纸本)9781665488280
Withthe immense computing power at our disposal, the numerical solution of partial differential equations (PDEs) is becoming a day-to-day task for modern computational scientists. However, the complexity of real-life problems is such that tractable solutions do not exist. this makes it difficult to validate the numerically obtained solution, so good error estimation is crucial in such cases. It allows the user to identify problematic areas in the computational domain that may affect the stability and accuracy of the numerical method. Such areas can then be remedied by either h- or p-adaptive procedures. In this paper, we propose to estimate the error of the numerical solution by solving the same governing problem implicitly and explicitly, using a different approximation order in each case. We demonstrate the newly proposed error indicator on the solution of a synthetic two-dimensional Poisson problem with tractable solution for easier validation. We show that the proposed error indicator has good potential for locating areas of high error.
暂无评论