Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore this 15-minute USENIX conference talk presenting HaSiS, a novel hardware-assisted single-index store designed specifically for Hybrid Transactional and Analytical Processing (HTAP). Learn how researchers from The Chinese University of Hong Kong, Shandong University, Indiana University Bloomington, and Rensselaer Polytechnic Institute address the fundamental conflict between optimal storage formats for OLTP and OLAP workloads. Unlike conventional multi-index approaches that struggle with trade-offs between performance and data freshness, discover how this paradigm-shifting single-index design leverages computational storage drives (CSDs) with built-in transparent compression to effectively serve HTAP workloads while achieving near-instant OLAP data freshness. The presentation covers the development of an HTAP-oriented B+-tree design and shares performance results comparing this CSD-assisted approach to state-of-the-art solutions.
