In the past two years the ATLAS Collaboration at the LHC has collected a large volume of data and published a number of ground breaking papers. The grid-based ATLAS distributedcomputing infrastructure played a crucia...
详细信息
In the past two years the ATLAS Collaboration at the LHC has collected a large volume of data and published a number of ground breaking papers. The grid-based ATLAS distributedcomputing infrastructure played a crucial role in enabling timely analysis of the data. We will present a study of the performance and usage of the ATLAS grid as platform for physics analysis in 2011. This includes studies of general properties as well as timing properties of user jobs (wait time, run time, etc). These studies are based on mining of data archived by the PanDA workload management system.
grid and Peer-to-Peer (P2P) networks enable the sharing and aggregation of geographically distributed resources. grids mostly deals with large computational problems and provide geographically distributed resources fo...
详细信息
ISBN:
(纸本)9781424402199
grid and Peer-to-Peer (P2P) networks enable the sharing and aggregation of geographically distributed resources. grids mostly deals with large computational problems and provide geographically distributed resources for large-scale data-intensive applications that generate large data sets, whereas P2P provides a way for sharing huge volumes of data. In a modern scientific computing, the communities of researchers involves in organizing, moving, visualizing, and analyzing massive amounts of a very large data collections in a geographically distributed environment. Research in the area of grid and P2P computing;have given us various ideas and solutions to address the problems. One of the most challenging issues in gridcomputing that address above requirements is locating and sharing remote data. In this paper, we illustrate the idea of using P2P approach for locating and sharing data in scientific data grid environment through our P2P Scientific Data grid Framework.
Performing delegation in large scale, dynamic and distributed environments with large numbers of shared resources is more challenging than inside local administrative domains. In dynamic environments like grids, on on...
详细信息
ISBN:
(纸本)9781424415595
Performing delegation in large scale, dynamic and distributed environments with large numbers of shared resources is more challenging than inside local administrative domains. In dynamic environments like grids, on one hand, delegating a restricted set of rights reduces exposure to attack but also limits the flexibility and dynamism of the application;on the other hand, delegating all rights provides maximum flexibility but increases exposure. This issue has not yet been adequately addressed by current grid security mechanisms and is becoming a very challenging and crucial issue for future grid development. Therefore, providing an effective delegation mechanism which meets the requirements of the least privilege principle is becoming an essential need. Furthermore, we are witnessing a phenomenal increase in the automation of organizational tasks and decision making, as well as the computerization of information related services, requiring automated delegation mechanisms. In order to meet these requirements we introduce an Active Delegation Framework which extends our previous work on on-demand delegation, making it context-aware. The framework provides a just-in-time, restricted and dynamic delegation mechanism for grids. In this paper we describe the development of this framework and its implementation and integration with the Globus Toolkit.
In this paper, we present a local-informationbased self-optimizing routing protocol Zigzag in virtual grid networks. A virtual grid network is obtained by virtually dividing a wireless network into a grid of geographi...
详细信息
ISBN:
(纸本)9780769550008
In this paper, we present a local-informationbased self-optimizing routing protocol Zigzag in virtual grid networks. A virtual grid network is obtained by virtually dividing a wireless network into a grid of geographical square regions called cells, and is used in MANETs and sensor networks to reduce energy consumption. A single node is selected as a router at each cell and inter-cell communication is realized by using the routers. Other nodes in the cell have no responsibility for inter-cell communication and can become inactive to save energy consumption. We consider maintenance of an inter-cell communication path to a destination node from its source node. When the destination node moves to a cell next to the current one, the path can be simply updated by extending it to the next cell. But, if the destination node moves around the network, the path becomes redundantly long and needs to be shortened. In this paper, we propose a self-optimizing routing protocol Zigzag in virtual grid networks, which can transform any given inter-cell path to a shortest (or minimum-hop) one by repeatedly applying local updates on the path. The routers locally and asynchronously update the path based only on local information and require no global information of the path such as the locations of the destination and the source nodes. We also show that the convergence time to a shortest path from any given path P is O(|P|) in the synchronous execution where |P| is the length (or the number of hops) of P.
Basic numerical synopses normal, least, whole were adequate for the business issues of the 1980s and 1990s. A lot of complex information, however, requires new systems. Perceiving client inclinations requires investig...
详细信息
ISBN:
(纸本)9781509036691
Basic numerical synopses normal, least, whole were adequate for the business issues of the 1980s and 1990s. A lot of complex information, however, requires new systems. Perceiving client inclinations requires investigation of procurement history, yet, likewise a nearby examination of searching conduct and items saw, remarks furthermore, audits signed on a site, and even grumblings and issues raised with client bolster staff. Foreseeing conduct requests that clients be assembled by their inclinations, so that conduct of one individual in the gathering can be utilized to anticipate the conduct of others. The calculations included incorporate characteristic dialect handling, example acknowledgment, machine learning and that's only the tip of the iceberg. These methods run extremely well on Hadoop. In this research we are propose PCP Predict crime pattern approach in hadoop for detect the pattern from the distributed warehouse.
Cloud computing, as a newly emerging technology, is an innovation providing dynamically scalable and virtualized resources as services. In this paper, we introduce our recent effort to build a service-oriented distrib...
详细信息
ISBN:
(纸本)9781424452781
Cloud computing, as a newly emerging technology, is an innovation providing dynamically scalable and virtualized resources as services. In this paper, we introduce our recent effort to build a service-oriented distributed computational system based on the Cloud concepts named distributed Computational Service Cloud. This kind of Cloud hosts scalable grid services, which are implemented with Web-Services-Resource-Framework-compliant (WSRF) [1] Web services, enabling high-performance and distributedcomputing. We evaluate the Cloud using scalable decision tree service, which provides computational intensive data mining algorithm. The architecture of the system as well as details of the distributed decision tree construction is the kernel content of this paper.
With the emergence of gridcomputing and service-oriented architectures, computing is becoming increasingly less confined to traditional computing platforms. gridcomputing promises the accessibility of vast computing...
详细信息
ISBN:
(纸本)0769524087
With the emergence of gridcomputing and service-oriented architectures, computing is becoming increasingly less confined to traditional computing platforms. gridcomputing promises the accessibility of vast computing and data resources across geographically dispersed areas. This capability is significantly enhanced by establishing support for mobile wireless devices to deliver access to high-performance computing under demanding circumstances. Access to the grid from mobile devices can be very effective in business environments where users can access the vast computing power and data repositories on the grid while working out on the field. This also enables and encourages collaborative working environments. A grid demonstrator system for distributed aircraft health monitoring already developed and implemented in the UK E-Science grid project, DAME, is introduced In this demonstrator, CBR technology for decision support is implemented in a practical framework that enables grid-based, proactive mobile computing.
gridcomputing using idle desktop PCs is popular as a low-cost platform for high-throughput applications. However, they require high-cost data transfer systems from file servers to calculating nodes when very large am...
详细信息
ISBN:
(纸本)9780889866386
gridcomputing using idle desktop PCs is popular as a low-cost platform for high-throughput applications. However, they require high-cost data transfer systems from file servers to calculating nodes when very large amounts of data are handled. This high cost derives from inexpensive I/O of worker PCs and small inexpensive file systems of desktop gridcomputing. This work describes a desktop gridcomputing system and proposes a resource management system for volume visualization. The resource management system proposed in this work overcomes problems with high-cost data transfer and helps us to use desktop gridcomputing for data-intensive applications. The resource management system provides efficient scheduling while considering the data file location and consists of three technical parts: resource management, multiple data replication, and worker selection. The proposed system uses these techniques to make an intelligent reuse of data which has been previously copied. This paper evaluates the proposed system compared with a traditional system that does not consider data file location. The proposed resource management System provides efficient scheduling and appears to achieve good performance for data-intensive applications.
Many distributed software systems allow participation by large numbers of untrusted, potentially faulty components on an open network. As faults are inevitable in this setting, these systems utilize redundancy and rep...
详细信息
ISBN:
(纸本)9780769543642
Many distributed software systems allow participation by large numbers of untrusted, potentially faulty components on an open network. As faults are inevitable in this setting, these systems utilize redundancy and replication to achieve fault tolerance. In this paper, we present a novel "smart" redundancy technique called iterative redundancy, which ensures efficient replication of computation and data given finite processing and storage resources, even when facing Byzantine faults. Iterative redundancy is more efficient and more adaptive than comparable state-of-the-art techniques that operate in environments with unknown system resource reliability. We show how systems that solve computational problems using a network of independent nodes can benefit from iterative redundancy. We present a formal analytical analysis and an empirical analysis, demonstrate iterative redundancy on a real-world volunteer-computing system, and compare it to existing methods.
The worlds' energy demand keeps increasing. To address this demand of energy growth intelligent energy management, monitoring and controlling play a big role besides increasing the energy generation capacity. The ...
详细信息
ISBN:
(纸本)9781479965571
The worlds' energy demand keeps increasing. To address this demand of energy growth intelligent energy management, monitoring and controlling play a big role besides increasing the energy generation capacity. The future smart grid expected to accommodate and integrates widely dispersed energy sources and distributed energy generations. Advanced information and communication technologies, smart meters and sensors are implemented in smart grid to enable an automated, intelligent and widely distributed energy delivery network. A huge amount of row data is collected by smart meters and sensors from the end user and different part of the network to the computation system, monitoring and controlling centers. An enormous pool of computing resources and storage must be provided to compute this vast amount of data. Processing these data enables the operators to make a knowledge based action as it occurs. This paper discusses the monitoring and controlling of distributed renewable energy generations in smart grid using cloud computing resources. Design and develop a Lab demonstrator and realization of the proposed solution to utilize the advantage of unlimited resources provided by cloud computing.
暂无评论