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

Abstract

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

On the Necessity of Explicit Cross-Layer Data Formats in Near-Data Processing Systems

“…First, we will give an overview over our system stack and discuss, how different parts of the stack interact with each other. Additionally, differences between this work and [27,28] will be highlighted.…”

“…Distributed and Parallel Databases (2022) 40: [27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45] the kernel or file system and distort the results somewhat. However, with a small on-device index cache, the NDP performance is around 33% better than the conventional stack.…”

“…Due to the under-utilization of the PL-portion, this opens up additional potential for hardware acceleration. Examples for this are given in [27,28]. In both works, the COSMOS hardware design is extended using processing elements that carry out computational tasks.…”

“…In both works, the COSMOS hardware design is extended using processing elements that carry out computational tasks. This section will discuss the performance of [27] as well as its applicability to typical use-cases.…”

Hardware acceleration of compression and encryption in SAP HANA

“…Ringlein et al [68] show the advantages (cost reduction, tail-latency minimization, execution efficiency) of using FPGA-based Function-as-a-Service into disaggregated cloud infrastructures, and Zha and Li [78] build an abstraction framework for virtualizing heterogeneous cloud FPGAs. Key-value stores also benefit from FPGAs with new memcached architectures [11], native computational storage [74], and hardware-managed transactions [36]. Kara et al [41] show how FPGAs can be used for coupling column-store ML algorithms with on-the-fly data transformation, such as decryption and delta-encoding decompression, while Lasch et al [43] accelerate Re-Pair compression algorithm using FPGAs.…”