README

What is Hydra?

Hydra is a high level, declarative language for modelling and simulation of physical systems. Systems in Hydra are being modelled using implicitly formulated (undirected) Differential Algebraic Equations (DAEs). While, physical modelling is our main focus any domain is fine where problems can be formulated using DAEs.

The language provides constructs for definition and composition of model fragments that enable modelling and simulation of large, complex, hybrid and highly structurally dynamic systems.

The first outline of Hydra was given by Nilsson et al. in the framework called Functional Hybrid Modelling (FHM). Subsequently, a number of papers has been published about the design and implementation of Hydra that can be accessed on the following web pages:

• http://www.cs.nott.ac.uk/~ggg/

• http://www.cs.nott.ac.uk/~nhn/

The implementation of Hydra

Currently, Hydra is implemented as a domain specific language embedded in Haskell. Hydra is still in active development.

How to get Hydra

The latest development version of Hydra can be obtained from its darcs repository by issuing the following command:

darcs get http://patch-tag.com/r/Hydra/pullrepo Hydra

You can also browse the sources in your browser: http://patch-tag.com/r/Hydra/snapshot/current/content/pretty

How to install Hydra

In order to install Hydra you need to have GHC (http://www.haskell.org/ghc/) version 6.10 (or later) available on your system. In addition, you need LLVM (http://llvm.org/) version 2.6.

The implementation of Hydra is packaged using Cabal. So it should be straight forward to install. Simply issue the following commands in the directory that you have just obtained using darcs:

runhaskell Setup.hs configure runhaskell Setup.hs build runhaskell Setup.hs install

If you have cabal-install tool available on your system there is an even simpler way. Simply issue the following command in the directory of Hydra:

cabal install

How to use Hydra

Currently, there is no documentation provided, only examples, I am afraid. There are also research papers (see above) that I recommend to read if you are interested. In any case, you are very welcome to contact me.

Cheers, George Giorgidze Email: ggg (at) cs.nott.ac.uk