“Version 5.6 continues our focus on delivering the significant performance and feature improvements our users have come to expect,” said Robert Bixby, Gurobi CEO.
Houston, TX (PRWEB) October 02, 2013
Gurobi Optimization has released version 5.6 of its leading mathematical programming solver, with significant performance improvements, new distributed optimization features, and several additional new capabilities users have been asking for.
“Version 5.6 continues our focus on delivering the significant performance and feature improvements our users have come to expect,” said Robert Bixby, Gurobi CEO. “Beyond double digit performance improvements when using a single machine, in version 5.6 we are adding significant new features to help users leverage the power of multiple machines, locally or in the cloud, to tune performance and more efficiently solve models.”
Improved LP and MIP Performance:
Performance testing using Gurobi’s test library, consisting of literally thousands of models, shows significant improvements when measured over a broad range of models in the set (models that require more than one second to solve), and even larger improvements for harder models (those that require more than 100 seconds to solve).
Specifically, for Linear Programming models, V5.6 delivers a 12% mean improvement in barrier performance for models in the broad test set and a 21% mean improvement for harder models. For Mixed Integer Programming models, V5.6 delivers a 14% mean improvement in time to proven optimality for models in a broad test set and a 29% mean performance improvement for harder models.
New Distributed Tuning Capabilities:
In V5.5, Gurobi introduced a new automatic parameter tuning tool to help users make more effective parameter choices with less effort. With V5.6 we are extending that capability to allow users to explore tuning options for a model not just on one machine, but on as many machines as they have available. This enables a dramatic increase in the amount of tuning exploration possible within a limited amount of time. On a broad set of models, given the same amount of tuning time we found that five machines produced settings that doubled the already significant improvements obtained on a single machine.
New Distributed Concurrent Optimization Capabilities:
Gurobi’s new distributed concurrent optimization capability allows users to harness the power of multiple Gurobi Compute Servers to solve a single MIP model more quickly. Each machine works independently, trying different settings in order to introduce additional diversity into the MIP search. By bringing the resources of multiple machines to bear on a single model, this approach yields significant performance improvements. We measured a 72% mean performance improvement over our broad set of difficult models when using five machines, as compared to a single machine.
In addition to the above enhancements, in V5.6, we’ve also responded to user requests for the following features: a) support for R 3.0 and Python 3.2, b) the ability to read and write MPS and LP files from within MATLAB, c) additional user control over the strategies Gurobi uses to solve models with SOS constraints as well as models with multiple, disconnected components, and d) additional control of asynchronous optimization.
About Gurobi Optimization, Inc.:
Gurobi (http://www.gurobi.com) is in the business of helping companies solve their hardest problems by providing the best optimization solver possible, outstanding support, and no surprises pricing. The Gurobi Optimizer is a state-of-the-art solver for linear programming (LP), quadratic programming (QP), quadratically constrained programming (QCP), mixed-integer linear programming (MILP), mixed-integer quadratic programming (MIQP), and mixed-integer quadratically constrained programming (MIQCP). It was designed from the ground up to exploit modern architectures and multi-core processors, using the most advanced implementations of the latest algorithms. Founded in 2008, Gurobi Optimization is based in Houston, TX (+1 713 871 9341).