The Alpaca (Cactus Tools for Application Level Performance and Correctness Analysis) project is aiming at developing complex, collaborative scientific applications, appropriate for highly scalable hardware architectures, providing fault tolerance, advanced debugging, and transparency against new developments in communication, programming, and execution models. Such tools are especially rare at the application level, where they are most critically needed.

In more details, the Alpaca project is to develop, at the application level:

(i) new fault tolerant capabilities that will be needed for increasingly large scale machines, where system and hardware faults are likely to be much more common than they are at present;

(ii) new performance monitoring capabilities, which will make it much easier for application developers and users to determine how their more complex application codes perform on current and future hardware;

(iii) new interactive debugging capabilities, critical to locate and cure software or algorithmic errors in suchcomplex, collaboratively written applications (which may be developed by global collaborations)

(iv) integration with Eclipse, the increasingly popular open source code development environment, making it much easier to develop code

(v) Alpaca tools will be developed with full involvement from application developers across a broad range of areas.

This project has the potential to have positive and dramatic impact on many disciplines, as Cactus is increasingly used by numerous groups in many application areas (including astrophysics, bio-informatics, CFD, and others), within the CS community, and by industry groups. It also plays a prominent role in numerous responses to NSF’s Track 2 and Track 1 solicitations. The proposed project will help groups develop codes Today that have a future migration path to systems and architectures of the future petascale environment.

The Alpaca project is described in more detail in a CCT Technical Report.