A parallel quasi-Newton method for partially separable large scale minimization

The parallel quasi-Newton method based on updating conjugate subspaces can be very effective for large-scale sparse minimization because conjugate subspaces with respect to sparse Hessians are usually easy to obtain. The authors demonstrate this point in this paper for the partially separable case with matrices updated by a quasi-Newton scheme of Griewank and Toint. The algorithm presented is suitable for parallel computation and economical in computer storage. Some testing results of the algorithm on an Alliant FX/8 minisupercomputer are reported.