17 December 2024 in ACM Transactions on Reconfigurable Technology and Systems
This article introduces DANSEN , the hardware accelerator component for neoDBMS, a full-stack computational storage system designed to manage on-device execution of database queries/transactions as a Near-Data Processing (NDP)-operation. The proposed system enables Database Management Systems (DBMS) to offload NDP-operations to the storage while maintaining control over data through a native storage interface . DANSEN provides an NDP-engine that enables DBMS to perform both low-level database tasks, such as performing database administration, as well as high-level tasks like executing SQL, on the smart storage device while observing the DBMS concurrency control. Furthermore, DANSEN enables the incorporation of custom accelerators as an NDP-operation (e.g., to perform hardware-accelerated ML inference directly on the stored data). We built the DANSEN storage prototype and interface on an UltraScale+HBM FPGA and fully integrated it with PostgreSQL 12. Experimental results demonstrate that the proposed NDP approach outperforms software-only PostgreSQL using a fast off-the-shelf NVMe drive and significantly improves the end-to-end execution time of an aggregation operation (similar to Q6 from CH-benCHmark, 150 million records) by ≈ 10.6×. The versatility of the proposed approach is also validated by integrating a compute-intensive data analytics application with multi-row results, outperforming PostgreSQL by ≈ 1.5×.
4 April 2025 in IEEE Transactions on Computers
Near-Data Processing (NDP) has been proven useful to accelerate Database Management Systems (DBMS) that handle infrequently accessed data stored in slow persistent storage. A key challenge for such an architecture is the synchronization of host-based and NDP operations, which require fine-grained interactions especially when the NDP device can also update (modify) the DBMS data autonomously. This paper introduces CINDA, the first full-stack computational storage capable of accelerating both read and update (write) database transactions using NDP. The proposed system relies on a hybrid host-device interface to enable the DBMS accessing persisted data, offloading computation to the storage device, and coordinating concurrent device-update operations with the host-update ones. A hybrid interface utilizes a cache-coherent interconnect such as CCIX or CXL for low-latency synchronization using a shared-lock table, and PCIe DMA for high-throughput bulk I/O. We evaluated the effectiveness of the proposed approach in a CCIX-based system by realizing an FPGA-based NDP-capable computational storage device and customizing an NDP-capable DBMS based on PostgreSQL to support update NDP operations. Our full-stack evaluation using the YCSB benchmark demonstrates that CINDA can deliver $\approx$4.2$\times$ end-to-end speedup when executing long-running update transactions directly on the storage device, while the host DBMS performs frequent short updates.
6 March 2026 in ACM Transactions on Database Systems
The performance and scalability of modern data-intensive systems processing large datasets are limited by unnecessary data movement. Even though near-data processing (NDP) can provably reduce data transfers and increase performance, at present, NDP is utilized primarily in read-only settings. Near-data execution of data-intensive modification operations is currently infeasible due to the lack of transactional consistency and the absence of practicable low-latency synchronization mechanisms between the host database engine and the NDP-engine on smart storage. In this article, we introduce update NDP as an approach to offloading modifications to computational storage with transactional guarantees in an NDP database system called neoDBMS . To ensure consistency, we introduce a low-latency shared lock table between the host and computational storage, based on novel cache-coherent interconnects . We also introduce a novel locking protocol that seamlessly integrates the shared lock table within the lock manager of the host NDP-engine. To handle failure recovery, while preserving high and robust performance, we introduce novel extended locking and logging mechanisms that allow the host and computational storage to perform useful work during log-movement. Our evaluation indicates that in-storage modifications in neoDBMS in mixed workload settings are ≥ 6.52× faster than host-only executions and exhibit robust performance due to lower data movement and better resource utilization.