Highlights of Gurobi Optimizer 2.0
- Significant improvements in MIP performance: For difficult models, Gurobi 2.0 is roughly four times as fast as Gurobi 1.1.
- Significant improvements in simplex performance: The new dual simplex optimizer is roughly twice as fast as Gurobi 1.1.
- Improvements in parallel MIP performance: Gurobi 2.0 extracts more benefit from each additional core in your multi-core system.
- MIP IIS (Irreducible Infeasible Subsystem): Gurobi 2.0 can identify the causes of infeasibility in MIP as well as LP models.
- MIP node files: Allow you to store MIP nodes on disk and thus solve much larger MIP models.
- Cloud computing: Gurobi is now available on-demand via the Amazon Elastic Computing Cloud.
- New examples: An extensive new set of examples illustrates key features and uses for Gurobi in all supported languages.
- Support for Mac OS X.
Gurobi Roadmap
The next major release - Gurobi Optimizer 3.0 - is targeted for April 2010. It will include a new parallel barrier solver for LP models, and continued performance improvement in MIP. Future releases, beyond Gurobi Optimizer 3.0, will include:
- Simplex QP (quadratic programming) solvers
- Barrier-based QP, and SOCP (second order cone programming) solvers
- MIQP (mixed-integer QP) and MISOCP (mixed-integer SOCP) solvers