The Swarm64 Data Accelerator (S64 DA): Processing OLAP Workloads Of Open-Source SQL-Databases With CPU+FPGA Cooperative Computing
Online Analytical Processing (OLAP) of very large data sets and/or high-velocity data is a workload that strains all parts of a compute system: storage bandwidth, IO-subsystem throughput, main-memory bandwidth, instruction-level concurrency and thread-parallelism. Swarm64 seeks to improve the effective throughput and the compute efficiency of OLAP workloads by adding FPGAs as additional compute element to standard compute servers. The hard- and software stack performing OLAP workloads, the S64 DA, has been designed to integrate into popular SQL open-source databases through standard interfaces (foreign data wrapper or storage engine). All that is required to move workloads to the S64 DA for processing, is to create fact-tables using a specific syntax and a specific type of index (the “optimised columns”). Once the tables have been created, no more changes to existing SQL-code are required. This talk provides insights into the concepts and technologies employed by Swam64 to speed-up OLAP workloads.
Karsten Rönner joined Swarm64 in 2014, about a year after the company’s foundation, and served first as a board member and chairman of the board. A year later he also took on the role of CEO. Swarm64 is the sixth start-up in which he has been involved as co-founder. Karsten has a background in computer architecture and decades of global experience in various management roles. He holds a master-degree in physics and a Ph.D. in electrical engineering, both from Leibniz University of Hannover, Germany.