DL_POLY_2 [1] is a package of subroutines, programs and data files, designed to facilitate molecular dynamics simulations of macromolecules, polymers, ionic systems and solutions on a distributed memory parallel computer. The first version was written on behalf of CCP5 [2] in the eighteen months preceding September 1993 when the (beta) release appeared. The principal authors at that stage were Bill Smith and Tim Forester at Daresbury Laboratory. For want of a more imaginative acronym the package was named DL_POLY. The entire DL_POLY project was given a major boost by the provision of a grant in January 1993 from the SERC's Computational Science Initiative. An additional grant was made by the EPSRC in June 1994 to continue developments in the period 1995-97. In April 1995 the project came under the Council for the Central Laboratory of the Research Councils.
There were many reasons for writing the package. The first reason being to exploit parallel computing for molecular simulation in the UK. There were several worthy packages available to the academic community including GROMOS [3], AMBER [4] and X-PLOR [5], but none of these were designed (as far as we were able to tell) with parallel computers in mind, (though parallel versions of these programs now exist). We decided then that the effort involved in devising a new code which had parallelism built into it from the start would be well spent. This we thought would allow us to adapt the code to new architectures with greater ease. The second reason was to bring to fruition the high level of parallel programming expertise at Daresbury Laboratory, and generally within the CCP5 community, in the area of molecular dynamics. We therefore thought it worthwhile to tap into this expertise to produce a new package. The third reason was that we thought it would be valuable to produce a package that was entirely free from commercial constraints, by which we mean the source code would be available to academic institutions for adaptation and extension. There would be no restriction preventing users from examining (and verifying!) the source code. DL_POLY was not to be a ``black box''.
Of course, in a rapidly developing subject area, we cannot guarantee we have everything a potential user may need. However since this is very much a project endorsed by CCP5, which serves the UK (and World-Wide) academic community, we are also hopeful that our users will be inclined to contribute any extensions of the package they devise and so accelerate its development. As it stands, the package has some nice features we believe. While it is designed for distributed memory parallel machines, we have taken care to ensure that it can, with minimum modification, be run on the popular workstations. Scaling up a simulation from a small workstation to a massively parallel machine remains straightforward.
Users are reminded that we are interested in hearing what other features could be usefully incorporated. We obviously have ideas of our own and CCP5 strongly influences developments, but other input would be welcome nevertheless. We also request that our users respect the integrity of DL_POLY_2 source and not pass it on to third parties. We require that all users of the package register with us, not least because we need to keep everyone abreast of new developments and discovered bugs. We have developed various forms of licence, which we hope will ward off litigation (from both sides), without denying access to genuine scientific users.
In the next section we outline the capabilities of DL_POLY_2 as briefly as possible.
This is followed by a description of the DL_POLY_2 directory structure and how to obtain
the source code from Daresbury Laboratory. Much more information is to be found later in
this manual. Access to all this information and more can be gained via our WWW site:
http://www.cse.clrc.ac.uk/msi/software/DL_POLY
.