VIOS: A Variation-Aware I/O Scheduler for Flash-Based Storage Systems

详细信息    查看全文

• 关键词：Process variation ; Solid state drive ; I/O scheduling ; Flash memory ; Parallelism
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9966
• 期：1
• 页码：3-16
• 全文大小：712 KB
• 参考文献：1.Min, C., Kim, K., Cho, H., Lee, S.W., Eom, Y.I.: SFS: random write considered harmful in solid state drives. In: USENIX Conference on File and Storage Technologies, p. 12, February 2012
  2.Hu, Y., Jiang, H., Feng, D., Tian, L., Luo, H., Ren, C.: Exploring and exploiting the multilevel parallelism inside SSDs for improved performance and endurance. IEEE Trans. Comput. 62(6), 1141–1155 (2013)MathSciNet CrossRef
  3.Chen, F., Lee, R., Zhang, X.: Essential roles of exploiting internal parallelism of flash memory based solid state drives in high-speed data processing. In: IEEE 17th International Symposium on High Performance Computer Architecture, pp. 266–277, February 2011
  4.Jung, M., Wilson III, E.H., Kandemir, M.: Physically addressed queueing (PAQ): improving parallelism in solid state disks. ACM SIGARCH Comput. Architect. News 40(3), 404–415 (2012)CrossRef
  5.Gao, C., Shi, L., Zhao, M., Xue, C.J., Wu, K., Sha, E.H.: Exploiting parallelism in i/o scheduling for access conflict minimization in flash-based solid state drives. In: 30th Symposium on Mass Storage Systems and Technologies, pp. 1–11, June 2014
  6.Li, P., Wu, F., Zhou, Y., Xie, C., Yu, J.: AOS: adaptive out-of-order scheduling for write-caused interference reduction in solid state disks. In: Proceedings of the International MultiConference of Engineers and Computer Scientists, vol. 1 (2015)
  7.Wang, M., Hu, Y.: An I/O scheduler based on fine-grained access patterns to improve SSD performance and lifespan. In: Proceedings of the 29th Annual ACM Symposium on Applied Computing, pp. 1511–1516, March 2014
  8.Jung, M., Choi, W., Srikantaiah, S., Yoo, J., Kandemir, M.T.: HIOS: a host interface I/O scheduler for solid state disks. ACM SIGARCH Comput. Architect. News 42, 289–300 (2014)CrossRef
  9.Ho, K.C., Fang, P.C., Li, H.P., Wang, C.Y.M., Chang, H.C.: A 45 nm 6b/cell charge-trapping flash memory using LDPC-based ECC and drift-immune soft-sensing engine. In: IEEE International Solid-State Circuits Conference Digest of Technical Papers, pp. 222–223, February 2013
  10.Zuloaga, S., Liu, R., Chen, P.Y., Yu, S.: Scaling 2-layer RRAM cross-point array towards 10 nm node: a device-circuit co-design. In: IEEE International Symposium on Circuits and Systems, pp. 193–196, May 2015
  11.Pan, Y., Dong, G., Zhang, T.: Error rate-based wear-leveling for NAND flash memory at highly scaled technology nodes. IEEE Trans. Very Large Scale Integration (VLSI) Syst. 21(7), 1350–1354 (2013)CrossRef
  12.Woo, Y.J., Kim, J.S.: Diversifying wear index for MLC NAND flash memory to extend the lifetime of SSDs. In: Proceedings of the Eleventh ACM International Conference on Embedded Software, p. 6, September 2013
  13.Shi, L.H., Di, Y., Zhao, M., Xue, C.J., Wu, K., Sha, E.H.M.: Exploiting process variation for write performance improvement on NAND flash memory storage systems. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 99, 1–4 (2015)
  14.Roh, H., Park, S., Kim, S., Shin, M., Lee, S.W.: B+-tree index optimization by exploiting internal parallelism of flash-based solid state drives. In: Proceedings of the VLDB Endowment, pp. 286–297 (2011)
  15.Kim, J., Oh, Y., Kim, E., Choi, J., Lee, D., Noh, S.H.: Disk schedulers for solid state drivers. In: Proceedings of the Seventh ACM International Conference on Embedded Software, pp. 295–304, October 2009
  16.Park, S., Shen, K.: FIOS: a fair, efficient flash I/O scheduler. In: USENIX Conference on File and Storage Technologies, p. 13, February 2012
  17.Shen, K., Park, S.: FlashFQ: a fair queueing I/O scheduler for flash-based SSDs. In: USENIX Annual Technical Conference, pp. 67–78, June 2013
  18.Hu, Y., Jiang, H., Feng, D., Tian, L., Luo, H., Zhang, S.: Performance impact and interplay of SSD parallelism through advanced commands, allocation strategy and data granularity. In: Proceedings of the International Conference on Supercomputing, pp. 96–107, May 2011
  19.Narayanan, D., Thereska, E., Donnelly, A., Elnikety, S., Rowstron, A.: Migrating server storage to SSDs: analysis of tradeoffs. In: The European Conference on Computer Systems, pp. 145–158, April 2009
  20.UMass Trace Repository. http://​traces.​cs.​umass.​edu
  21.Cui, J., Wu, W., Zhang, X., Huang, J., Wang, Y.: Exploiting process variation for read and write performance improvement of flash memory. In: 32th International Conference on Massive Storage Systems and Technology, May 2016
  
• 作者单位：Jinhua Cui (18)
  Weiguo Wu (18)
  Shiqiang Nie (18)
  Jianhang Huang (18)
  Zhuang Hu (18)
  Nianjun Zou (18)
  Yinfeng Wang (19)
  
  18. School of Electronic and Information Engineering, Xi’an Jiaotong University, Shaanxi, 710049, China
  19. Department of Software Engineering, ShenZhen Institute of Information Technology, Guangdong, 518172, China
  
• 丛书名：Network and Parallel Computing
• ISBN：978-3-319-47099-3
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：9966

文摘

NAND flash memory has gained widespread acceptance in storage systems because of its superior write/read performance, shock-resistance and low-power consumption. I/O scheduling for Solid State Drives (SSDs) has received much attention in recent years for its ability to take advantage of the rich parallelism within SSDs. However, most state-of-the-art I/O scheduling algorithms are oblivious to the increasingly significant inter-block variation introduced by the advanced technology scaling. This paper proposes a variation-aware I/O scheduler by exploiting the speed variation among blocks to minimize the access conflict latency of I/O requests. The proposed VIOS schedules I/O requests into a hierarchical-batch structured queue to preferentially exploit channel-level parallelism, followed by chip-level parallelism. Moreover, conflict write requests are allocated to faster blocks to reduce access conflict of waiting requests. Experimental results shows that VIOS reduces write latency significantly compared to state-of-the-art I/O schedulers while attaining high read efficiency.