-
Classes Class Description no.uib.cipr.matrix.distributed.BlockDiagonalPreconditioner theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.CollectiveCommunications theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.Communicator theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.DistColMatrix theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.DistRowMatrix theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.DistVector theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.Reduction theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.Reductions theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.no.uib.cipr.matrix.distributed.TwoLevelPreconditioner theno.uib.cipr.matrix.distributedpackage has been deprecated because of a number of hard to fix concurrency bugs. It is distributed only for backwards compatibility, but is not recommended. The utility of this package is questionable, as it does not allow distribution of computation between JVMs or across a network. For many people, distributed computing of multiple matrices can be achieved at a user-level through the JPPF Framework. Users who need to deal with few very large matrices may wish to implement their own storage classes and solvers using JPPF, but this will not be supported directly in matrix-toolkits-java.