chemora2c.pdf (214.8 kB)
Chemora: A Programming Framework for Physics
The fundamental difficulty with scientific software
is that the existing languages are either too generalpurpose,
too low-level, or both. Thus, physics
equations are entangled with the way they are
implemented and optimized. Because everything is
coupled, it is difficult to adapt to algorithm or
hardware advances. More importantly, it means that
researchers cannot specialize as easily.
In principle, it should be possible to
independently specify (1) the scientific equations to
be solved; (2) the type of grids; (3) the numerical
methods, e.g. time integrators, elliptic solvers; (4)
the execution platform (desktop, accelerated cluster,
etc.); (5) performance goals, e.g. time, power, etc.