Grid and Cloud Computing and their Application
Main Article Content
Dear SCPE Reader,
We present a Special Issue on Grid and Cloud Computing and their Applications. The first six papers included in this issue are modified and extended versions of papers presented at the 8th International Conference on Parallel Processing and Applied Mathematics, PPAM 2009, which took place on September 13–16, 2009 in WrocÅaw, Poland. These papers have been selected from contributions originally accepted in the frame of two workshops: the 4th Grid Applications and Middleware Workshop, GAMW'2009 and the 5th Workshop on Large Scale Computations on Grids, LaSCoG'09. The remaining four papers were selected from the responses to an Open Call for Papers.
The selected papers are addressing a large variety of current research topics related to the call: distributed file systems, scheduling, load balancing, data mining, simulations, mobile agents, scientific portals, workflows, service negotiations or distributed visualization.
The first paper, VOFS: A Secure Churn-Tolerant Grid File System presents a specially designed secure file system that allows the members of a virtual organization to share files. A decentralized common file namespace is proposed to avoid a single point of failure. The proposed software stack includes a P2P system of file servers and can operate in a dynamic Grid environment
The second paper, Matching Jobs With Resources: an Application-Driven Approach proposes a distributed matchmaker, named GREEN, which provides Grid users with features for easy submission of job execution requests containing performance requirements. GREEN relies on a two-level benchmarking methodology: resources are characterized by means of their performance evaluated through the execution of low-level and application-specific benchmarks.
In the third paper, Prediction and Load Balancing System for Distributed Storage, the application of a common mass storage system model in a national distributed storage system has been described. The prediction and load balancing subsystem, which provides advanced monitoring functionalities is discussed. The proposed system makes use of replication techniques to increase availability and performance of data access.
The fourth paper Distributed Data Integration and Mining Using Admire Technology presents the data integration engine for environmental data. The proposed software is being developed in the scope of the ADMIRE project. The proposed platform allows for integration of data from distributed, heterogeneous resources. It also allows users to construct reusable application processing elements specified in a DMIL, a language for data mining and integration.
The fifth paper Ultra-Fast Carrier Transport Simulation on the Grid. Quasi-Random Approach studies quasi-random number generation in a Grid-enabled package named Stochastic ALgorithms for Ultra-fast Transport in sEmiconductors (SALUTE). The performance of the corresponding algorithms on the Grid is also discussed. A large number of tests are reported on the EGEE and the SEEGRID Grid infrastructures.
In the sixth paper, Management of High Performance Scientific Applications Using Mobile Agents Based Services, an explanation of how programmers can extend their applications to exploit services on heterogeneous and distributed platforms is provided. A native console is implemented, using mobile agents to control the application life-cycle. Moreover, software agents implement a mobile service that supports check-pointing, suspension, resuming, cloning and migration of managed applications.
The seventh paper Vine Toolkit—Towards Portal Based Production Solutions For Scientific and Engineering Communities With Grid-Enabled Resources Support addresses the challenge of synchronization of distributed workflows, and establishing a community driven Grid environment for the seamless results sharing and collaboration. The proposed toolkit offers user interface web components to be embedded in the existing portals, integration with a workflow engine, Grid security, and a built-in meta-scheduling mechanism allowing automatic load balancing among data centers to meet peak demands.
The eight paper Fast Multi-Objective Rescheduling of Workflows to Constrained Resources Using Heuristics and Memetic Evolution describes GORBA, a global optimising resource broker and allocator, which is designed to be used in a static planning environment. Several heuristics for rescheduling are introduced and their contribution to the overall planning process is studied.
The ninth paper VieSLAF Framework: Facilitating Negotiations in Clouds by Applying Service Mediation and Negotiation Bootstrapping presents a novel framework for the specification and management of service level agreement (SLA) mappings and meta-negotiations facilitating service mediation, negotiation and bootstrapping in Cloud computing environments. The users may specify, manage, and apply SLA mappings without a-priori knowledge about negotiation protocols, required security standards or negotiated terms.
Finally, the tenth paper Large Scale Problem Solving Using Automatic Code Generation and Distributed Visualization presents a new approach to solving four important scalability challenges: programming productivity, scalability to large numbers of processors, I/O bandwidth, and interactive visualization of large data. The approach uses the Cactus framework, automated code generation, and numerical methods. A demonstration of the proposed system was awarded first place in the IEEE SCALE 2009 Challenge.
We would like to express our gratitude to all referees who have worked to help authors to improve the quality of papers selected to be published in this Special Issue.
Special Issue Editors