Pattern operators are extensions to the Pat- tern/Skeleton parallel programming approach used to apply two types of communication patterns to the same data. The operators are intended to simplify the wide range of possi- ble patterns and skeletons. The abstraction helps manage non-functional concerns on the Grid/Cloud environments. This paper explains how the pattern operators work on synchronous cyclic undirected graph patterns, and it shows examples on how they are used. A prototype was created to test the feasibility of the idea. The example used to show the operator approach is the addition of termination detection to a discrete solution to a PDE. The example can be coded with 27.31% less non-functional code than a similar implementation in MPJ, and its programmability index is 13.5% compared to MPJ's 9.85%. The overhead for an empty pattern with low communication was 15%. The use of pattern operators can reduce the number of skeletons/patterns developed.
The primary concern in proffering an infrastructure for general purpose computational grids formation is security. Grid implementations have been devised to deal with the security concerns. The chief factors that can be problematic in the secured selection of grid resources are the wide range of selection and the high degree of strangeness. Moreover, the lack of a higher degree of confidence relationship is likely to prevent efficient resource allocation and utilization. In this paper, we propose an efficient approach for the secured selection of grid resources, so as to achieve secure execution of the jobs. The presented approach utilizes trust and reputation for securely selecting the grid resources by also evaluation user’s feedback on the basis of the feedback already available about the entities. The proposed approach is scalable for an increased number of resources.
Grid networks take advantage of the resources available by many computers connected within the network to solve large-scale computational problems. To achieve a scalable and reliable Grid network system, the workload needs to be efficiently distributed among the computing resources accessible on the network. Therefore, a distributed and scalable load-balancing scheme for Grid Networks is proposed in this paper. Another objective of this paper is to develop a latency reduction workload distribution protocol for Grid networks. Here, we demonstrate that introducing a latency reduction factor in the random sampling can reduce the effects of communication latency in the Grid network environment. Simulation results show that the resulted network system provides an effective, scalable, and reliable load-balancing scheme for the distributed resources available on Grid networks.
Scientific computing often requires the availability of a massive number of computers for performing large scale experiments. Traditionally, high-performance computing solutions and installed facilities such as clusters and super computers have been employed to address these needs. Cloud computing provides scientists with a completely new model of utilizing the computing infrastructure with the ability to perform parallel computations using large pools of virtual machines (VMs). The infrastructure services (Infrastructure-as-a-service), provided by these cloud vendors, allow any user to provision a large number of compute instances. However, scientific computing is typically characterized by complex communication patterns and requires optimized runtimes. Today, VMs are manually instantiated, configured and maintained by cloud users. These coupled with the latency, crash and omission failures in service providers, results in an inefficient use of VMs, increased complexity in VM-management tasks, a reduction in the overall computation power and increased time for task completion. In this paper, a high performance cloud computing strategy is proposed that combines the adaptation of a parallel processing framework, such as the Message Passing Interface (MPI) and an efficient checkpoint infrastructure for VMs, enabling its effective use for scientific computing. By developing such a mechanism, we can achieve optimized runtimes comparable to native clusters, improve checkpoints with low interference on task execution and provide efficient task recovery. In addition, check pointing is used to minimize the cost and volatility of resource provisioning, while improving overall reliability. Analysis and simulations show that the proposed approach compares favorably with the native cluster MPI implementations.
There has been growing interest in the use of mobile devices for providing access to the various applications, services and resources within grid computing environments. As is the case for most access to grid resources, the use of mobile devices requires user authentication. Given the limited amount of power available for mobile devices, the applications on these devices, including the authentication services used to access grid resources, must be designed to conserve power. In this paper, we present an authentication architecture utilizing a “lightweight” user-centric authentication approach. Specifically, we aim to provide mobile users with access to advance resource reservation within grid environments, and, consequently, our approach incorporates this objective. The approach, however, applies to user authentication with any grid resource or service. In attempting to overcome the limitations of mobile devices, such as limited battery power, mobile users can utilize grid environments in a transparent and secure way.
