I'm a Junior Researcher at Group for Theoretical & Computational Physics, Research Center for Physics, Indonesian Institute of Sciences (http://teori.fisika.lipi.go.id). Research activities in my group fall into three areas: theoretical physics, computational and computing sciences. I have been mostly involved in the area of computing science, especially in developing computing tools using parallel environment.

Thereafter I am assigned to lead a small team to establish a reliable but low cost parallel computing system since the end of 2006. This is the first one at the Indonesian Institute of Sciences, and one of few dedicated clusters in Indonesia. In contrast to the conventional clusters, we have modified the whole architecture and make it as an open infrastructure provided for public with full-ownership policy. This is mostly intended to provide a real training field for parallel programmers in Indonesia to perform any parallel computation and all aspects of parallel environment as well, without having the parallel machine itself and through an integrated and user-friendly web interface. Moreover, it is provided for free.

Since its grand-launching in 2007, the facility has enjoyed numerous potential users, not only the students but also some small research groups. These facts have proven the feasibility of this kind of clusters which is the first approach in the world. Although the approach is motivated by the lack infrastructure in the developing countries like Indonesia, it leads to some innovations which might be applicable for parallel environment with special purposes around the world.

The cluster is named as LIPI Public Cluster (LPC) and accessible at http://www.cluster.lipi.go.id. We have now released the whole system under GNU Public License (GPL) at SourceForge (http://sourceforge.net/projects/openpc/).

This year we have further developed a web-service based middleware to make the blocks of nodes owned by users in LPC to be connectable to the global grid with various grid middleware. Once the middleware has been installed successfully, our users would have their own cluster machines and are able to connect them to their global partners under certain grid infrastructures. Again, this is globally a new approach to improve the public-accessibility and usability of cluster facilities.

On the other hand, my group also conducts research topics on computational science and data mining using LPC. Currently, since 2007 I have initiated a project on “indirect” topical data integration system using the so-called “focused web-harvesting”. We have deployed a new approach using pre-defined web-harvesting to realize an automated and intelligent data collection available on the web, and further restructure the data to be searchable in smarter ways. The approach is appropriate to overcome the problem of integrating data from multi sources without forcing the participating institutions to follow a particular standard.

We now enter the last (fourth) phase of development, but the system, named as Indonesian Scientific Index (ISI), is already fully running and accessible for public at http://www.isi.lipi.go.id. Again, it is released under GPL at SourceForge (http://sourceforge.net/projects/openisi/). During the last phase we are now developing an automated data reconstruction of huge number of collected data, and then perform comprehensive content analysis to obtain relevant knowledge’s and indicators in a particular topic.