发明名称 Efficient parallel processing of a network with conflict constraints between nodes
摘要 According to one exemplary embodiment, a method for parallel processing a network of nodes having at least one ordering constraint and at least one conflict constraint is provided. The method may include breaking a plurality of loops caused by the at least one ordering constraint. The method may also include determining a node order based on the at least one ordering constraint. The method may then include determining a conflict order based on the at least one conflict constraint, whereby no new loops are created in the network. The method may further include performing parallel processing of the network of nodes based on the node order and the conflict order.
申请公布号 US9608868(B2) 申请公布日期 2017.03.28
申请号 US201615267166 申请日期 2016.09.16
申请人 International Business Machines Corporation 发明人 Gupta Hemlata;Hathaway David J.;Kalafala Kerim;Rose Ronald D.
分类号 G06F15/16;H04L12/24;H04L12/26 主分类号 G06F15/16
代理机构 代理人 Bunker Robert C.
主权项 1. A computer program product for parallel processing a network of nodes having at least one ordering constraint and at least one conflict constraint having a coupling between a pair of nets, comprising: one or more non-transitory computer-readable storage medium and program instructions stored on at least one of the one or more non-transitory tangible storage medium, the program instructions executable by a processor, the program instructions comprising: program instructions to determine the at least one ordering constraint by mapping at least one input connection and at least one output connection to each node within the network of nodes, wherein the at least one ordering constraint indicates that the input to a node must be processed before the output from the node; program instructions to break a plurality of loops caused by the at least one ordering constraint using a loop cutting algorithm; program instructions to determine a node order based on the at least one ordering constraint, wherein the node order comprises: (i) a partial node order based on a levelization scheme assigning level values to each node within the network of nodes, wherein a successor node is assigned a greater level value and a lower level value is assigned to a preceding node level within the network of nodes and (ii) a strict node order based on determining an estimated processing cost for an endpoint node associated with the at least one conflict constraint; program instructions to determine a conflict order based on the at least one conflict constraint, wherein no new loops are created in the network, and wherein the conflict order comprises determining an estimated processing cost for an endpoint node associated with the at least one conflict constraint, and wherein the estimated processing cost comprises an apportioned processing cost of a plurality of downstream nodes that are downstream of the endpoint node; and program instructions to perform parallel processing of the network of nodes based on the node order and the conflict order using multi-threaded dynamic scheduling, wherein the parallel processing includes noise analysis of an electronic network.
地址 Armonk NY US