Advertisement

TL: A High Performance Buffer Replacement Strategy for Read-Write Splitting Web Applications

  • Zhiwen Jiang
  • Yong Zhang
  • Jin Wang
  • Chao Li
  • Chunxiao Xing
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8709)

Abstract

With the quickly increasing pressure of users’ requests, more and more Web applications adopt the read-write splitting to improve the performance, which requires new buffer replacement strategy. Flash memory has emerged as a popular storage media, and we use it to replace hard disk for read-write splitting Web applications. Traditional buffer replacement strategies are suboptimal on flash memory, since flash memory has some distinguished features as out-of-place update and read-write asymmetry. In this paper, we design a flash-aware buffer replacement strategy Tri-List (TL), which emphasizes the I/O asymmetry of flash memory. Unlike other flash-aware buffer replacement policies which are based on LRU method, TL considers both recency and frequency of page requests to make better decision of buffer replacement. Experimental results on both synthetic and benchmarking traces shows that TL has up to 30.3% improvements than state-of-the-art flash-aware buffer management policies.

Keywords

Buffer Size Flash Memory Buffer Management Solid State Drive Buffer Pool 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bouganim, L., Jonsson, B.T., Bonnet, P.: uFlip: Understanding flash IO patterns. In: CIDR (2009)Google Scholar
  2. 2.
    Chou, H.-T., DeWitt, D.J.: An evaluation of buffer management strategies for relational database systems. In: VLDB (1985)Google Scholar
  3. 3.
    Do, J., Zhang, D., Patel, J.M., DeWitt, D.J., Naughton, J.F., Halverson, A.: Turbocharging DBMS buffer pool using SSDs. In: SIGMOD Conference, pp. 1113–1124 (2011)Google Scholar
  4. 4.
    Do, J., Zhang, D., Patel, J.M., DeWitt, D.J.: Fast peak-to-peak behavior with SSD buffer pool. In: ICDE, pp. 1129–1140 (2013)Google Scholar
  5. 5.
    Jiang, S., Zhang, X.: LIRS: an efficient low inter-reference recency set replacement policy to improve buffer cache performance. In: SIGMETRICS, pp. 31–42 (2002)Google Scholar
  6. 6.
    Johnson, T., Shasha, D.: 2Q: A low overhead high performance buffer management replacement algorithm. In: VLDB 1994, pp. 439–450 (1994)Google Scholar
  7. 7.
    Jung, H., Shim, H., Park, S., Kang, S., Cha, J.: LRU-WSR: Integration of LRU and writes sequence reordering for flash memory. IEEE Transactions on Consumer Electronics 54(3), 1215–1223 (2008)CrossRefGoogle Scholar
  8. 8.
    Kim, H., Ahn, S.: BPLRU: A buffer management scheme for improving random writes in flash storage. In: FAST (2008)Google Scholar
  9. 9.
    Li, Z., Jin, P., Su, X., Cui, K., Yue, L.: CCF-LRU: A new buffer replacement algorithm for flash memory. IEEE Transactions on Consumer Electronics 55(3), 1351–1359 (2009)CrossRefGoogle Scholar
  10. 10.
    Lv, Y., Cui, B., He, B., Chen, X.: Operation-aware buffer management in flash-based systems. In: SIGMOD Conference 2011, pp. 13–24 (2011)Google Scholar
  11. 11.
    O’Neil, E.J., O’Neil, P.E., Weikum, G.: The LRU-K page replacement algorithm for database disk buffering. In: SIGMOD Conference 1993, pp. 297–306 (1993)Google Scholar
  12. 12.
    Ou, Y., Haerder, T.: Clean first or dirty first? A cost-aware self-adaptive buffer replacement policy. In: IDEAS, pp. 7–14 (2010)Google Scholar
  13. 13.
    Ou, Y., Haerder, T., Jin, P.: CFDC: A flash-aware replacement policy for database buffer management. In: DaMoN (2009)Google Scholar
  14. 14.
    Seo, D., Shin, D.: Recently-evicted-first buffer replacement policy for flash storage devices. IEEE Transactions on Consumer Electronics 54(3), 1228–1235 (2008)CrossRefMathSciNetGoogle Scholar
  15. 15.
    Robinson, J.T., Devarakonda, M.V.: Data cache management using frequency-based replacement. In: SIGMETRICS, pp. 134–142 (1990)Google Scholar
  16. 16.
  17. 17.
  18. 18.
    Park, S.Y., Jung, D., Kang, J.-U., et al.: CFLRU: a replacement algorithm for flash memory. In: CASES, pp. 234–241 (2006)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Zhiwen Jiang
    • 1
  • Yong Zhang
    • 1
  • Jin Wang
    • 1
  • Chao Li
    • 1
  • Chunxiao Xing
    • 1
  1. 1.RIIT, TNList, Department of Computer Science and Technology, Center for High-speed Railway TechnologyTsinghua UniversityBeijingChina

Personalised recommendations