Need for a deeper cross-layer optimization for dense NAND SSD to improve read performance of big data applications: A case for melded pages

Arpith, K and Gopinath, K (2020) Need for a deeper cross-layer optimization for dense NAND SSD to improve read performance of big data applications: A case for melded pages. In: HotStorage 2020 - 12th USENIX Workshop on Hot Topics in Storage and File Systems, co-located with USENIX ATC 2020, 13-14 July 2020, Virtual, Online.

	PDF hot_2020_12th-usenix.pdf - Published Version Restricted to Registered users only Download (277kB) | Request a copy
Preview	PDF PPT.pdf - Published Version Download (872kB) | Preview

Abstract

In the case of dense NAND flash such as TLC, the LSB, CSB and MSB1 pages in a wordline can be combined to form a larger logical page called melded-page. In this paper, we propose melding TLC/QLC pages to improve the performance of SSD by mitigating the overheads involved in the read operation. Melded-pages are read in their entirety. The controller schedules the write requests in such a way that, during reads later, requests for data present in LSB, CSB and MSB pages are guaranteed to be present in the request queue. This method works reliably when the workload performs its read operations in large chunks or has a predictable I/O pattern. We obtain performance improvements of up to 45 on workloads that use large block sizes such as the Hadoop Distributed File System (HDFS). Big data solutions that exhibit such read patterns can vastly benefit from melding pages. © HotStorage 2020 - 12th USENIX Workshop on Hot Topics in Storage and File Systems, co-located with USENIX ATC 2020. All rights reserved.

Item Type:	Conference Paper
Publication:	HotStorage 2020 - 12th USENIX Workshop on Hot Topics in Storage and File Systems, co-located with USENIX ATC 2020
Publisher:	USENIX Association
Additional Information:	The copyright of the article belongs to the USENIX Association
Keywords:	Big data; File organization; NAND circuits, Big data applications; Cross layer optimization; Data solutions; Hadoop distributed file system (HDFS); Large blocks; NAND Flash; Read operation; Read performance, Digital storage
Department/Centre:	Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	02 Dec 2020 11:37
Last Modified:	09 Dec 2022 06:20
URI:	https://eprints.iisc.ac.in/id/eprint/67022

Actions (login required)

View Item