Invited Session Wed.1.H 1058

Wednesday, 10:30 - 12:00 h, Room: H 1058

Cluster 10: Implementations & software [...]

Implementations of interior point methods for linear and conic optimization problems


Chair: Erling Dalgaard Andersen



Wednesday, 10:30 - 10:55 h, Room: H 1058, Talk 1

Csaba Meszaros
Exploiting hardware capabilities in implementations of interior point methods


The talk concerns the implementation of interior point methods for solving large-scale optimization problems. In our investigation we focus on the exploitation of the recently introduced AVX vector instruction set and show that the capabilities of modern processors can be highly exploited by special implementation techniques. We describe the implementation design of our interior point solver and demonstrate that its performance on standard multi-core platforms can reach 100 Gflops when solving large-scale optimization problems.



Wednesday, 11:00 - 11:25 h, Room: H 1058, Talk 2

Erling Dalgaard Andersen
On recent improvements in the interior-point optimizer in MOSEK


In this talk we will discuss the recent advances in the interior-point optimizer in the upcoming version 7 release of MOSEK.
The advances include better dense column handling, an improved GP ordering for the normal equations, handling of intersection cones and
warmstart capabilities. Beyond these advances the interior-point optimizer has also been extended to handle semi-definite optimization problems.



Wednesday, 11:30 - 11:55 h, Room: H 1058, Talk 3

Imre Polik
Crossing over

Coauthor: Philipp M. Christophel


There are only few academic papers about crossover techniques, i.e., about algorithms that take an optimal solution of an LP and "round'' it to an optimal basic solution. Moreover, the problem we face in practice is very different from the setup in these papers. In this talk we wish to highlight these differences and offer new techniques for the different problems. Besides the standard case of interior-point methods, other issues we are discussing are solutions from the network simplex method, basic infeasibility and unboundedness certificates, and perturbation techniques.
Computational experiments using SAS/OR will be presented.


  personal loans. Therefore, we can say that the active substances in its composition are more perfectly mixed. Vardenafil is not only present in the original Buy Levitra, but also as part of its analogs.