Publication at Faculty of Mathematics and Physics |
2020
Abstract
The problem of data distribution in NoSQL databases is par-ticularly difficult in the case of graph databases since the data oftenrepresent a large, highly connected graph. We face this task with mon-itoring of user queries, for which we created a logging module provid-ing information serving as an input to a redistribution algorithm whichbases on a lightweight method of Adaptive Partitioning but incorporatesour enhancements overcoming its present drawbacks (local optima, bal-ancing, edge weights).
The results of our experiments show 70% - 80%reduction of communication between cluster nodes which is a compara-ble result to other methods, which, however, are more computationallydemanding or suffer from other shortcomings.