Scippy

UG

Ubiquity Generator framework

About

UG is a generic framework to parallelize branch-and-bound based solvers (e.g., MIP, MINLP, ExactIP) in a distributed or shared memory computing environment.

  • Exploits powerful performance of state-of-the-art "base solvers", such as SCIP, CPLEX, etc.
  • Without the need for base solver parallelization

Base solvers and communication libraries are abstracted within UG. A parallel solver instantiated by UG framework is named:

  • ug[Base solver, Communication libaray]

News

29/Feb/2016 Beta version UG 0.8.2 is released as a part of the SCIP Optimization Suite 3.2.1.
31/Jul/2015 Patched version UG 0.8.1 is released, replacing UG 0.8.0 of the SCIP Optimization Suite 3.2.0.
01/Jul/2015 Beta version UG 0.8.0 is released as a part of the SCIP Optimization Suite 3.2.0.
18/Dec/2014 Beta version UG 0.7.5 is released as a part of the SCIP Optimization Suite 3.1.1.
27/Feb/2014 Beta version UG 0.7.3 is released as a part of the SCIP Optimization Suite 3.1.0.
25/Feb/2014 Website relaunched.
04/Jan/2013 Beta version UG 0.7.1 is released as a part of the SCIP Optimization Suite 3.0.1.
01/Aug/2012 Beta version UG 0.7.0 is released as a part of the SCIP Optimization Suite 3.0.0.


NOTE

ug[SCIP,Pth] (FiberSCIP) needs SCIP (and SoPlex) library compiled with "PARASCIP=true" option.

License

UG is distributed under the ZIB Academic License. You are allowed to retrieve UG for research purposes as a member of a non-commercial and academic institution.
If you want to use UG commercially or if you are interested in maintenance and support, please contact us by sending an email to Thorsten Koch.

See ZIB Academic Licence for more information.

Any publication for which UG is used must include an acknowledgment and a reference to the article:

Yuji Shinano, Tobias Achterberg, Timo Berthold, Stefan Heinz, and Thorsten Koch,
ParaSCIP – a parallel extension of SCIP,
Competence in High Performance Computing 2010, editors: C. Bischof, H.-G. Hegering, W. E. Nagel, and G. Wittum, pages 135–148, Springer, 2012.

Here is the corresponding bibtex entry.

Download

The latest version of ParaSCIP and FiberSCIP is included in the SCIP Optimization Suite. Please refer to the provided README file for help on installing the software.
It is highly recommended to install the patched version UG-0.8.1 instead of the included version of UG.

Projects

Current projects are as follows:

ug[SCIP,MPI]
(ParaSCIP)
  • Uses CIP solver SCIP
  • Used to investigate a large scale parallelization
  • Runs on up to 7,168 cores at supercomputer HLRN II
  • Runs on up to 512 cores on a supercomputer at ISM
  • Computed optimal solutions for ds, stp3d from MIPLIB2003
  • Computed optimal solutions for 50-10v, probportfolio, reblock354, rmatr200-p20, dg012142, dc1c, germany50-DBM from MIPLIB2010
ug[SCIP,Pthreads]
(FiberSCIP)
  • Uses CIP solver SCIP
  • Enables parallelization on single desktop computers
  • Evaluated by Hans Mittelmann (9/Aug/2012)
ug[CPLEX,MPI]
(ParaCPLEX)

ParaSCIP and FiberSCIP can handle both MIPs and MINLPs and are a parallel framework for solving mixed integer (linear and nonlinear) programs.They have deterministic mode for debugging.

Reference

Yuji Shinano, Tobias Achterberg, Timo Berthold, Stefan Heinz, Thorsten Koch, ParaSCIP -- a parallel extension of SCIP.
Yuji Shinano, Stefan Heinz, Stefan Vigerske, Michael Winkler, FiberSCIP -- A shared memory parallelization of SCIP.

Cooperation

Research by using UG is in cooperation with: The development of UG is supported by Google Research Grant and the DFG Research Center Matheon, in particular within Project B20.