Method and system for optimizing data storage of query statistics of graph database

A computer device comprising a memory and a processor, wherein the processor is configured to execute non-transitory computer instructions for optimizing data storage of query statistics of a graph database, comprising a query server, a meta server, and storage servers;

• wherein the meta server periodically scans, on the storage servers in which partitions are located, all edges in the partitions;
• the storage server determines, according to all the edges in the partitions, partitions to which start points and end points belong, and calculates outgoing-edge correlation and incoming-edge correlation between partitions;
• and
• the storage server calculates relevancies between partitions through a preset correlation matrix weight according to the outgoing-edge correlation and the incoming-edge correlation between partitions, wherein partitions with high relevancies are stored on a same storage server;
• wherein the calculating outgoing-edge correlation and incoming-edge correlation between partitions comprises:
• calculating the outgoing-edge correlation between partitions based on the following formula:
• OUT
• ij
• =
• Δ
• {
• out_edge
• ⁢
• _count
• i
• j
• out_total
• ⁢
• _edge
• i
• ⁢
• if
• ⁢
• i
• ≠
• j
• 1
• ⁢
• if
• ⁢
• i
• =
• j
• ,
• wherein OUTij represents outgoing-edge correlation between partition i and partition j, out_total_edgei represents a total number of outgoing edges with start points in partition i, and out_edge_countij represents the number of outgoing edges with start points located in partition i and end points located in partition j;
• and
• calculating the incoming-edge correlation between partitions based on the following formula:
• IN
• ij
• =
• Δ
• {
• in_edge
• ⁢
• _count
• i
• j
• in_total
• ⁢
• _edge
• i
• ⁢
• if
• ⁢
• i
• ≠
• j
• ;
• 1
• ⁢
• if
• ⁢
• i
• =
• j
• ,
• wherein INij represents incoming-edge correlation between partition i and partition j, in_total_edgei represents a total number of incoming edges with end points in partition i, and in_edge_countij represents the number of incoming edges with start points located in partition j and end points located in partition i;
• wherein the calculating relevancies between partitions through a preset correlation matrix weight according to the outgoing-edge correlation and the incoming-edge correlation between partitions comprises:
• calculating the relevancies between partitions according to the outgoing-edge correlation and the incoming-edge correlation between partitions and the preset correlation matrix weight based on the following formula:
• Rij=αOUTij+(1−α)INij, wherein α is the preset correlation matrix weight, Rij represents a relevancy between partition i and partition j, OUTij represents outgoing-edge correlation between partition i and partition j, and INij represents incoming-edge correlation between partition i and partition j.