Managed acceleration for In-Memory database analytic workloads. Issue 4 (4th July 2017)

• Record Type: Journal Article
• Authors: O'Neill, Eoghan; McGlone, John; Kilpatrick, Peter; Nikolopoulos, Dimitrios
• Is Part Of: International journal of parallel, emergent and distributed systems. Volume 32:Issue 4(2017)
• Page Start: 406
• Page End: 427
• Publication Date: 2017-07-04
• Subjects: OpenCL -- heterogeneous computing -- workload management -- In-Memory database -- predictive analysis; Parallel computers -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; Computer algorithms -- Periodicals; 004.35
• DOI: 10.1080/17445760.2016.1170832
• Languages: English
• ISSNs: 1744-5760

Abstract

Abstract : In-Memory Databases (IMDBs), such as SAP HANA, enable new levels of database performance by removing the disk bottleneck and by compressing data in memory. The consequence of this improved performance means that reports and analytic queries can now be processed on demand. Therefore, the goal is now to provide near real-time responses to compute and data intensive analytic queries. To facilitate this, much work has investigated the use of acceleration technologies within the database context. While current research into the application of these technologies has yielded positive results, they have tended to focus on single database tasks or on isolated single user requests. This paper uses SHEPARD, a framework for managing accelerated tasks across shared heterogeneous resources, to introduce acceleration into an IMDB. Results show how, using SHEPARD, multiple simultaneous user queries all receive speed-up by using a shared pool of accelerators. Results also show that offloading analytic tasks onto accelerators can have indirect benefits for other database workloads by reducing contention for CPU resources. Graphical Abstract: Interaction of the SHEPARD management framework to process user queries from an in-memory database system.