Journal of the Korea Society of Computer and Information (한국컴퓨터정보학회논문지)

Korean Society of Computer Information (한국컴퓨터정보학회)
권호 표지
DOI QR코드

DOI QR Code

Hot Data Verification Method Considering Continuity and Frequency of Write Requests Using Counting Filter

  • Lee, Seung-Woo (Dept. of Computer Science, Kyungpook National University) ;
  • Ryu, Kwan-Woo (Dept. of Computer Science, Kyungpook National University)
  • Received : 2019.03.18
  • Accepted : 2019.06.05
  • Published : 2019.06.28
Download PDF
⟨ Previous Next ⟩

Abstract

Hard disks, which have long been used as secondary storage in computing systems, are increasingly being replaced by solid state drives (SSDs), due to their relatively fast data input / output speeds and small, light weight. SSDs that use NAND flash memory as a storage medium are significantly different from hard disks in terms of physical operation and internal operation. In particular, there is a feature that data overwrite can not be performed, which causes erase operation before writing. In order to solve this problem, a hot data for frequently updating a data for a specific page is distinguished from a cold data for a relatively non-hot data. Hot data identification helps to improve overall performance by identifying and managing hot data separately. Among the various hot data identification methods known so far, there is a technique of recording consecutive write requests by using a Bloom filter and judging the values by hot data. However, the Bloom filter technique has a problem that a new bit array must be generated every time a set of items is changed. In addition, since it is judged based on a continuous write request, it is possible to make a wrong judgment. In this paper, we propose a method using a counting filter for accurate hot data verification. The proposed method examines consecutive write requests. It also records the number of times consecutive write requests occur. The proposed method enables more accurate hot data verification.

Keywords

CPTSCQ_2019_v24n6_1_f0001.png 이미지

Fig. 1. Flash Memory based Storage System

CPTSCQ_2019_v24n6_1_f0002.png 이미지

Fig. 2. Bloom filter

CPTSCQ_2019_v24n6_1_f0003.png 이미지

Fig. 3. Counting filter

CPTSCQ_2019_v24n6_1_f0004.png 이미지

Fig. 4. Multi Hash Function Framework

CPTSCQ_2019_v24n6_1_f0005.png 이미지

Fig. 5. Bloom filter Hash Function Framework

CPTSCQ_2019_v24n6_1_f0006.png 이미지

Fig. 6. operation process of the proposed method

CPTSCQ_2019_v24n6_1_f0007.png 이미지

Fig. 7. continuity of write request

CPTSCQ_2019_v24n6_1_f0008.png 이미지

Fig. 8. Judgment Process 1

CPTSCQ_2019_v24n6_1_f0009.png 이미지

Fig. 9. Judgment Process 2

CPTSCQ_2019_v24n6_1_f0010.png 이미지

Fig. 10. CPU clock cycle comparison

CPTSCQ_2019_v24n6_1_f0011.png 이미지

Fig. 11. Number of hot data pages when applying Bloom filter technique and Counting filter technique

CPTSCQ_2019_v24n6_1_f0012.png 이미지

Fig. 12. Write requests with short continuity

CPTSCQ_2019_v24n6_1_f0013.png 이미지

Fig. 13. Number of hot data pages when applying MHF technique and CMHF

Table 1. Rate of positive error according to bit vector size and number of hash functions

CPTSCQ_2019_v24n6_1_t0001.png 이미지

Table 2. Trace file characteristics

CPTSCQ_2019_v24n6_1_t0002.png 이미지

References

  1. Gartner, https://www.gartner.com/en/newsroom/pressreleases/2017-10-17-gartner-says-worldwide-device-shipments-will-increase-2-percent-in-2018
  2. Tae-Sun Chung, Dong-Joo Park, Dong-Ho Lee, Sang-Won Lee, and Ha-Joo Song, "System Software for Flash Memory: A Survey", 2004.
  3. Jun Liu, Shuyu Chen, Tianshu Wu, Hancui Zhang, "A Novel Hot Data Identification Mechanism for NAND Flash Memory," IEEE Journals & Magazines, Volume: 61, Issue: 4 pp.463-469, 2015.
  4. Jen-Wei Hsieh, Tei-Wei Kuo, Li-Pin Chang, "Efficient identification of hot data for flash memory storage systems", ACM Transactions on Storage (TOS), Volume 2 Issue 1, February 2006.
  5. Hyun-Seob Lee, Hyun-Sik Yun, and Dong-Ho Lee, "HFTL:Hybrid Flash Translation Layer based on Hot Data Identification for Flach Memory", IEEE Journals & Magazines, 2009.
  6. Dong-chul Park, David H.C Du, "Hot Data Identification for Flash-based Storage Systems Using Multiple Bloom Filters", 2011 IEEE 27th Symposium on Mass Storage Systems and Technologies (MSST).
  7. L.-P. Chang, "On efficient wear-leveling for largescale flash-memory storage systems," Proc. of the 2007 ACM symposium on Applied computing, pp.1126-1130, 2007.
  8. Hyesook Lim, Jungwon Lee, Changhoon Yim, "Complement Bloom Filter for Identifying True Positiveness of a Bloom Filter" IEEE Communications Letters ( Volume: 19 , Issue: 11 , Nov. 2015 )
  9. Ori Rottenstreich, Isaac Keslassy, "The Bloom Paradox: When Not to Use a Bloom Filter" IEEE/ACM Transactions on Networking ( Volume: 23 , Issue: 3 , June 2015 )
  10. Peizhen Lin, Feng Wang, Weiliang Tan, Hui Deng, "Enhancing Dynamic Packet Filtering Technique with d-Left Counting Bloom Filter Algorithm" 2009 Second International Conference on Intelligent Networks and Intelligent Systems
  11. Access Pattern, http://tech.kakao.com/2016/07/17/coding-for-ssd-part-5/