R/parallel is a project which objective is to bring closer parallel computing to any R user, with special interest on bioinformaticians daily tasks. To reduce the gap between users and technology, R/parallel main efforts are focused on providing a tool easy-to-use, useful with common analysis tasks, available to as many users as possible by reducing external dependencies on additional software and/or hardware, and efficient with currently available resources.
Detailed information and examples based on bioinformatics problems are provided in this web site.
The development of R/parallel is organized in three consecutive projects
with the following titles:
These projects will increase, one after another, the number of computer resources that will collaborate on parallel computations.
The first project, In my desktop, enables parallel computing in R using several processing units (e.g. processor cores) on a single desktop computer. The second subproject, In my office, increases the computing power by aggregating the unused cpu cycles of several networked computers, for example the workstations of a regular department or office. The final extension of R/parallel is achieved with the third project, Everywhere, which enables the collaboration of remote compute resources, for example, between long distant organizations, or enabling the use of existing dedicated clusters.
The R add-on package, an outcome of project 1, as well as additional information, is available at different sections of this web site.
After the recent publication of an article in
we consider that Project 1 has finalized, and we have moved forward to Project 2.
However, we are aware, thanks to the collaboration of many people interested in R/parallel, of several issues, bugs or new functionalities that we have to introduce with new releases, regardless of the ongoing project. One example is the porting of R/parallel to MacOS X. We are working on that.
For those expecting the release of Project 2, right now, we are planning the delivery of a first working version by the end of year.
Any questions or suggestions regarding R/parallel can be addressed to:
Gonzalo Vera Rodríguez
email: gonzalo.vera AT rparallel.org
This work is the result of the collaboration between the following research organizations: