In-memory data grid hash scheme optimization

摘要

Systems and methods of managing an in-memory data grid (IMDG) may involve conducting a data distribution analysis of the IMDG on a periodic basis, and selecting a hash scheme from a plurality of hash schemes based on the data distribution analysis. In one example, the selected hash scheme is used to conduct a repopulation of the IMDG, wherein the repopulation increases the distribution evenness of database records across the IMDG.

主权项

1. A computer implemented method comprising:
conducting a data distribution analysis of an in-memory data grid, the in-memory data grid comprising database content, wherein the data distribution analysis includes an analysis of the amount and type of database content stored by a plurality of virtual machines in the in-memory data grid; selecting one hash scheme from a plurality of hash schemes based on the data distribution analysis and information known about the plurality of hash schemes regarding the type of database content the hash schema is better suited, wherein the plurality of hash schemes includes at least a content-based scheme and a mathematical scheme, the mathematical scheme selected when said content-based scheme is less efficient at distributing the data than the mathematical scheme, wherein the mathematical scheme performs a mathematical operation on a search key and wherein the content-based scheme groups similar content together; and using only the selected hash scheme to conduct a repopulation of the in-memory data grid to a fixed number of partitions during a maintenance period, wherein the repopulation increases a distribution evenness of database records across the in-memory data grid and the selected hash scheme is used to access data from said in-memory data grid.