Near-data processing in database systems on native computational storage under HTAP workloads

Today's Hybrid Transactional and Analytical Processing (HTAP) systems, tackle the ever-growing data in combination with a mixture of transactional analytical workloads. While optimizing for aspects such as freshness performance isolation, they build on traditional data-to-code principle may trigger massive cold transfers that impair overall scalability. Firstly, this paper we show Near-Data (NDP) naturally fits HTAP design space. Secondly, propose an NDP database architecture, allowing transactionally consistent in-situ executions operations settings. We evaluate proposed architecture state-of-the-art key/value-stores multi-versioned DBMS. In contrast to setups, our approach yields robust, resource- cost-efficient performance.