[Help-glpk] Gassmann (2012) : Optimization services : connecting algebraic modelling languages to several solvers using a web-aware framework

[Robbie Morrison]

Hello all

The following shortly-to-be presented paper might be of interest.

Robbie
---

ISMP 2012
21st International Symposium on Mathematical Programming
19-24 August 2012, Berlin, Germany

abstracts :
http://www.ismp2012.org/images/stories/bookofabstracts_onlineversion.pdf

Friday 24-Aug-2012 10:30–12:00

Fri.1.H 1058

Open source software for modeling and optimization

Organizer/Chair: Theodore Ralphs, Lehigh University

Invited Session: Gus Gassmann, Dalhousie University
  (with Jun Ma, Kipp Martin)

Optimization services : connecting algebraic modelling
  languages to several solvers using a web-aware
  framework

The common paradigm in mathematical optimization uses a
modular approach consisting of an instance generator,
e.g., an algebraic modelling language (AML), and a
solver.  Loosely coupled systems allow the substitution
of one solver or AML for another.  This is especially
attractive when one considers open-source software,
such as the suite of solvers that make up the COIN-OR
project.  However, the communication of solver options
is an often overlooked detail.  Solver developers often
use options specific to their own solvers, and even
where two solvers use the same option, syntax and
interpretation may differ.  This can be cumbersome,
especially if the AML and solver reside on different
computers.  In addition, open-source solvers are often
layered on top of other solvers, which adds to the
complexity, since solver options may have to be
directed at different levels in the solver hierarchy.
Optimization Services is a web-aware framework that
provides a common interface between AMLs and a variety
of open-source and commercial solvers.  This talk
explores some of the difficulties encountered in
connecting the COIN-OR solver suite to a common AML,
and the methods we used to overcome them.

---
Robbie Morrison
PhD student -- policy-oriented energy system simulation
Institute for Energy Engineering (IET)
Technical University of Berlin (TU-Berlin), Germany
University email (redirected) : address@hidden
Webmail (preferred)           : address@hidden
[from Webmail client]