Large Scale Computations on Grids

Abstract

The annual Large Scale Computation on Grids (LaSCoG) Workshop, as well as the Scalable Computing: Practice and Experience (SCPE) Journal, attempt at providing interactive and professional forum for the distributed computing community. Therefore the SCPE was selected to publish the extended versions of the latest (2007) LaSCoG meeting.

The 3rd LaSCoG meeting was held in Gdansk, Poland, on September 11th, 2007, in conjunction with the Seventh International Conference on Parallel Processing and Applied Mathematics (PPAM). Of the 16 submissions received, 6 were chosen for delivery and inclusion in the PPAM post-proceedings. These papers were limited to ten pages, thus their authors were invited to publish extended versions of their contributions in the SCPE. After careful re-refereeing and further improvements, it is my pleasure to provide you with the following collection of papers.

The first paper, by Sunil John and John Morrison, introduces the object oriented condensed graph, a computational paradigm which combines condensed graphs model for expressing parallelism with object oriented model as an effective programming paradigm in order to be able to develop large scale parallel and distributed systems.

In the second paper, Peter Slížik and Ladislav Hluchý presents several components of the simulation models of natural disasters developed in the frame of the European project MediGrid.

The third paper, by Gabriel Rodríguez, Xoán C. Pardo, María J. Martín, Patricia González, and Daniel Díaz, describes the design and implementation of a service-based architecture named CPPC-G (Controller-Precompiler for Portable Checkpointing on Grid) designed to manage execution of fault tolerant applications on Grids.

In the fourth paper, Dana Petcu, Alexandru Cârstea, Georgiana Macariu, and Marc Frîncu describe an approach to integrate legacy computer algebra systems into a service-oriented architecture, namely the SymGrid-Services.

The next paper, by David E. Singh, Alejandro Miguel, Félix García, and Jesús Carretero discusses MASIPE, an agent-based tool for monitoring parallel applications in Grids.

The last paper from the LaSCoG Workshop is authored by Assel Akzhalova, Daniar Aizhulov, Galymzhan Seralin, and Gulnar Balakayeva, and describes a problem solving environment implemented as Web portal. Its application is to allow Grid utilization of numerical methods that use MPI for their parallel implementations.

Finally, the seventh paper, by Igor Rozman, Marjan Šterk, Jaka MoÄnik, Borut RobiÄ, and Roman Trobec, was submitted directly to this Special Issue. It analyzes communication and computational performance results on an ad-hoc computing network composed of geographically wide spread distributed nodes.

Note that while the above mentioned papers represent rather different topics, all of them are intensively studied in the last years and I hope that the reader of this issue will find the papers interesting for his or her future research activities.

I would also like to thank my co-Chair of LaSCoG—Prof. Dana Petcu—who has withdrawn herself from editing this Special Issue due to the fact that she has co-authored one of the included papers. In 2007 most work in preparation of LaSCoG was done by her and I am really grateful for all that she has done. I would also like to thank all members of the Program Committee, for their diligent work promoting LaSCoG and hard work refereeing papers. Thank you all!

Finally, it is my pleasure to invite all readers to take part in LaSCoG 2008, which will take place in WisÅa, Poland, on October 20–22, 2008. There is still time to contribute. More information can be found at: http://www.lascog.imcsit.org

Marcin Paprzycki