Impact of processor cache memory on storage performance
- Authors
- Kim, Young-Kuen; Song, Yong Ho
- Issue Date
- May-2018
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- Cache; Flash translation layer; NAND flash memory; Processor
- Citation
- Proceedings - International SoC Design Conference 2017, ISOCC 2017, pp.304 - 305
- Indexed
- SCOPUS
- Journal Title
- Proceedings - International SoC Design Conference 2017, ISOCC 2017
- Start Page
- 304
- End Page
- 305
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/5304
- DOI
- 10.1109/ISOCC.2017.8368908
- ISSN
- 0000-0000
- Abstract
- Recently flash memory provides high performance, high capacity, and stable quality of service (QoS). To satisfy these features, a complex Flash Translation Layer (FTL) algorithm is used, which greatly increases the amount of data processed by the processor. As a result, data traffic increases between processor and cache memory, resulting in more cache misses. Frequent cache misses cause cache miss penalty to degrade overall storage performance. In this paper, we analyze the effect of cache size on storage devices. As a result of verifying the performance on the FPGA board, when the cache size increased from 2KB to 16KB, the performance increase was 273.8% for random write, 214.4% for random read, 281.6% for sequential write, and 313.3% for sequential read.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 융합전자공학부 > 1. Journal Articles

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.