US 12,147,473 B2
Graph data processing method and apparatus, and system
Yinglong Xia, Santa Clara, CA (US); Jian Xu, Shenzhen (CN); and Mingzhen Xia, Hangzhou (CN)
Assigned to Huawei Technologies Co., Ltd., Shenzhen (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed on Jan. 20, 2022, as Appl. No. 17/580,605.
Application 17/580,605 is a continuation of application No. 16/426,287, filed on May 30, 2019, granted, now 11,256,749.
Application 16/426,287 is a continuation of application No. PCT/CN2017/112117, filed on Nov. 21, 2017.
Claims priority of application No. 201611081960.0 (CN), filed on Nov. 30, 2016.
Prior Publication US 2022/0147571 A1, May 12, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 15/16 (2006.01); G06F 9/48 (2006.01); G06F 9/50 (2006.01); G06F 16/27 (2019.01); G06F 16/901 (2019.01)
CPC G06F 16/9024 (2019.01) [G06F 9/4881 (2013.01); G06F 9/50 (2013.01); G06F 16/27 (2019.01)] 14 Claims
OG exemplary drawing
 
1. A graph data processing method, comprising:
obtaining, by a master node, graph data, wherein the graph data comprises M vertexes and a plurality of directional edges, each edge connects two vertexes of the M vertexes, a direction of each edge is from a source vertex to a destination vertex in the two vertexes, and M is an integer greater than two;
dividing, by the master node, the graph data, into P non-overlapping shards, wherein each of the P shards comprises at least one incoming edge directed to at least one vertex in the corresponding shard, a sum of quantities of vertexes corresponding to the P shards is equal to M, P is a positive integer greater than 1;
scheduling, by the master node, at least two edge sets comprised in a first shard of the P shards and an associate edge set comprised in a second shard of the P shards for processing by at least two worker nodes, wherein the associate edge set comprises an outgoing edge directed from a vertex in the first shard; and
synchronizing, by the master node, a first data block stored by a first worker node with a second data block stored by a second worker node based on setting a synchronization flag that instructs the first worker node to perform data synchronization with the second worker node after processing the first data block, wherein the first data block is scheduled last according to a scheduling sequence in the first shard, and wherein the second data block is scheduled last in data blocks corresponding to the associate edge set.