Comment by kingcauchy

How did overall memory footprint look during ingestion vs Redis? Have any tradeoff graphs as the dataset gets larger and with mixed workloads?