Description:
As of now, when using aggregations, we maintain a bucket for each distinct group produced by the group by clause of the aggregation, in memory. Due to this, memory usage depends on the number of such distinct groups.
When the number of distinct groups is not finite, memory consumption will not be finite as well.
In order to avoid this, we should avoid keeping these buckets in memory forever.
sybernix
commented
3 years ago
Description: As of now, when using aggregations, we maintain a bucket for each distinct group produced by the group by clause of the aggregation, in memory. Due to this, memory usage depends on the number of such distinct groups.
When the number of distinct groups is not finite, memory consumption will not be finite as well. In order to avoid this, we should avoid keeping these buckets in memory forever.
Related Issue: https://github.com/siddhi-io/siddhi/issues/1696