Github user ndwangsen commented on the issue:
https://github.com/apache/carbondata/pull/2314
@kumarvishal09 If the user specifies or default that the minimum data load
of the node is less than the average data amount of each node, the existing
strategy is used to handle---
