Class MultipleLook


  • public class MultipleLook
    extends DAXWriter
    This class ends up writing a partitioned dax, that corresponds to one partition as defined by the Partitioner. Whenever it is called to write out a dax corresponding to a partition it looks up the dax i.e parses the dax and gets the information about the jobs making up the partition.
    Version:
    $Revision$
    Author:
    Karan Vahi
    • Field Detail

      • mNodeSet

        private java.util.Set mNodeSet
        The set of job id's in the partition.
      • mRelationsMap

        private java.util.Map mRelationsMap
        A map containing the relations between the jobs making up the partition.
      • mPartADAG

        private org.griphyn.vdl.dax.ADAG mPartADAG
        The ADAG object containing the partitioned dax.
      • mNumOfJobs

        private int mNumOfJobs
        The number of jobs that are in the partition.
      • mCurrentNum

        private int mCurrentNum
        The number of jobs about which the callback interface has knowledge.
      • mIndex

        private int mIndex
        The index of the partition that is being written out.
    • Constructor Detail

      • MultipleLook

        public MultipleLook​(java.lang.String daxFile,
                            java.lang.String directory)
        The overloaded constructor.
        Parameters:
        daxFile - the path to the dax file that is being partitioned.
        directory - the directory in which the partitioned daxes are to be generated.
    • Method Detail

      • writePartitionDax

        public boolean writePartitionDax​(Partition partition,
                                         int index)
        It writes out a dax consisting of the jobs as specified in the partition.
        Specified by:
        writePartitionDax in class DAXWriter
        Parameters:
        partition - the partition object containing the relations and id's of the jobs making up the partition.
        index - the index of the partition.
        Returns:
        boolean true if dax successfully generated and written. false in case of error.