大数据存储系统I/O性能优化技术研究进展

doi:10.11959/j.issn.2096-0271.2017062

摘要/Abstract

摘要：

大数据存储系统的I/O性能是影响大数据应用整体性能的关键因素之一，总结了当前在存储系统架构、元数据I/O性能、数据I/O性能方面开展的大数据存储系统I/O性能优化工作，并指出了未来大数据存储系统I/O性能优化的一些研究方向。

关键词: 大数据存储系统, 存储系统架构, 元数据I/O性能, 数据I/O性能, 性能优化技术

Abstract:

The I/O performance of big data storage system is one of the key factors that affect the overall performance of big data applications.The I/O performance optimization of big data storage system in storage system architecture,metadata I/O performance and data I/O performance was summarized,and some research directions of I/O performance optimization for big data storage systems in future were pointed out.

Key words: big data storage system, storage system architecture, metadata I/O performance, data I/O performance, performance optimization technology

中图分类号:

TP399

肖利民, 霍志胜. 大数据存储系统I/O性能优化技术研究进展[J]. 大数据, 2017, 3(6): 65-84.

Limin XIAO, Zhisheng HUO. Review of I/O performance optimization technology for big data storage system[J]. Big Data Research, 2017, 3(6): 65-84.

图/表 1

参考文献 102

[8]	CHEN M HILDEBRAND D , NELSON H , et al. vNFS:maximizing NFS performance with compounds and vectorized I/O[C]// The 15th USENIX Conference on File and Storage Technologies, February 27 March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:301-314.
[9]	MADHYASTHA T M , REED D A . Learning to classify parallel input/output access patterns[J]. IEEE Transactions on Parallel ＆ Distributed Systems, 2002,13(8):802-813.
[10]	WANG Y J KAELI D . Profileguided I/O partitioning[C]// The 17th Annual International Conference on Supercomputing (ICS'03), June 23-26,2003, San Francisco,USA. New York: ACM Press, 2003:252-260.
[11]	PATRICK C M KANDEMIR M , KARAK O¨Y M , et al. Cashing in on hints for better prefetching and caching in PVFS and MPI-IO[C]// The 19th ACM International Symposium on High Performance Distributed Computing, June 21-25,2010, Chicago,USA. New York: ACM Press, 2010:191-202.
[12]	ALKISWANY S , GHARIBEH A , RIPEANU M . The case for a versatile storage system[J]. ACM SIGOPS Operating Systems Review, 2010,44(1):10-14.
[13]	NARAYAN S , CHANDY J A . ATTEST:attributes-based extendable storage[J]. Journal of Systems ＆ Software, 2010,83(4):548-556.
[14]	LI X , XIAO L , QIU M , et al. Enabling dynamic file I/O path selection at runtime for parallel file system[J]. The Journal of Supercomputing, 2014,68(2):996-1021.
[15]	KIM S KIM H , LEE J , et al. Enlightening the I/O path:a holistic approach for application performance[C]// The 15th USENIX Conference on File and Storage Technologies, February 27 March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:345-358.
[16]	LU Y , SHU J , WANG W , et al. ReconFS:a reconstructable file system on flash storage[C]// The 12th USENIX Conference on File and Storage Technologies, February 17-20,2014, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2014:75-88.
[17]	LEE E , BAHN H , NOH S H . Unioning of the buffer cache and journaling layers with non-volatile memory[C]// The 11th USENIX Conference on File and Storage Technologies, February 12-15,2013, San Jose,USA. Berkeley: USENIX Association Berkeley, 2013:73-80.
[18]	OH Y , CHOI J , LEE D , et al. Caching less for better performance:balancing cache size and update cost of flash memory cache in hybrid storage systems[C]// The 10th USENIX Conference on File and Storage Technologies, February 14-17,2012, San Jose,USA. Berkeley: USENIX Association Berkeley, 2012:25.
[19]	HUO Z S , XIAO L M , ZHONG Q L , et al. A metadata cooperative caching architecture based on SSD and DRAM for file systems[C]// The 15th International Conference on Algorithms and Architectures for Parallel Processing, November 18-20,2015, Zhangjiajie,China. Berlin: Springer, 2015:31-51.
[20]	YADGAR G , YAAKOBI E , SCHUSTER A , et al. Write once,get 50% free:saving SSD erase costs using WOM codes[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:257-271.
[21]	LEE C , SIM D , HWANG J , et al. F2FS:a new file system for flash storage[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:273-286.
[22]	LEE C , LIU M , JUN S , et al. Applicationmanaged flash[C]// The 14th USENIX Conference on File and Storage Technologies, February 22-25,2016, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2016:339-353.
[23]	ARTEAGA D , CABRERA J B , XU J , et al. CloudCache:on-demand flash cache management for cloud computing[C]// The 14th USENIX Conference on File and Storage Technologies, February 22-25,2016, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2016:355-369.
[24]	YAN S , LI H , HAO M , et al. Tiny-Tail Flash:near-perfect elimination of garbage collection tail latencies in NAND SSDs[C]// The 15th USENIX Conference on File and Storage Technologies, February 27 March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:15-28.
[25]	ZHANG S , CATANESE H , WANG A A , et al. The composite-file file system:decoupling the one-toone mapping of files and metadata for better performance[C]// The 14th USENIX Conference on File and Storage Technologies, February 22-25,2016, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2016:15-22.
[26]	KUMAR H , PATEL Y , KESAVAN R , et al. High performance metadata integrity protection in the WAFL copy-on-write file system[C]// The 15th USENIX Conference on File and Storage Technologies, February 27-March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:197-212.
[27]	JOHNSON C M , KEETON K , MORREY C B , et al. From research to practice:experiences engineering a production metadata database for a scale out file system[C]// The 12th USENIX Conference on File and Storage Technologies, February 17-20,2014, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2014:191-198.
[28]	THOMSON A , ABADI D J . CalvinFS:consistent WAN replication and scalable metadata management for distributed file systems[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:1-14.
[29]	NIAZI S , ISMAIL M , HARIDI S , et al. HopsFS:scaling hierarchical file system metadata using newSQL databases[C]// The 15th USENIX Conference on File and Storage Technologies (FAST'17), February 27-March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:89-104.
[30]	LEUNG A , ADAMS I , MILLER E L . Magellan:a searchable metadata architecture for large-scale file systems:UCSC-SSRC-09-07[R]. California:University of California, 2009.
[31]	LEUNG A W , SHAO M , BISSON T , et al. Spyglass:fast,scalable metadata search for large-scale storage systems[C]// The 7th Conference on File and Storage Technologies, February 24-27,2009, San Francisco,USA. Berkeley: USENIX Association Berkeley, 2009:153-166.
[32]	GIFFORD D K , JOUVELOT P , SHELDON M A , et al. Semantic file systems[J]. Symposium on Operating Systems Principles, 1991,25(5):16-25.
[33]	HUA Y , JIANG H , ZHU Y , et al. Smartstore:a new metadata organization paradigm with semantic-awareness for next-generation file systems[C]// The Conference on High Performance Computing Networking,Storage and Analysis, November 14-20,2009, Portland,USA. New Jersey: IEEE Press, 2009:1-12.
[34]	HUANG H H , ZHANG N , WANG W , et al. Just-in-time analytics on large file systems[J]. IEEE Transactions on Computers, 2012,61(11):1651-1664.
[35]	TAKATA M , SUTOH A . Eventnotification-based inactive file search for large-scale file systems[C]// APMRC, October 31-November 2,2012, Singapore. New Jersey: IEEE Press, 2012:1-7.
[36]	HUO Z S , XIAO L M , ZHONG Q L , et al. MBFS:a parallel metadata search method based on bloomfilters using MapReduce for large-scale file systems[J]. The Journal of Supercomputing, 2016,72(8):3006-3032.
[37]	GALEN S . 2015 Parallel file system requirements[R].[S.l.]: Lustre Scalability Workshop, 2009.
[38]	RODEH O , TEPERMAN A . zFS:a scalable distributed file system using object disks[C]// The 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies, April 7-10,2003, San Diego,USA. New Jersey: IEEE Press, 2003:20-27.
[39]	ZHU Y , JIANG H , WANG J , et al. HBA:distributed metadata management for large cluster-based storage systems[J]. International Journal of Engineering Innovations ＆ Research, 2008,19:750-763.
[1]	AVERY C , KENIN C , LI J , et al. Highperformance techniques for parallel I/O[M]. Boca Raton: CRC Press, 2007.
[2]	CHING A CHOUDHARY A , COLOMA K , et al. Noncontiguous I/O accesses through MPI-I/O[C]// The 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid, May 12-15,2003, Tokyo,Japan. New Jersey: IEEE Press, 2003:104-111.
[40]	BRANDT S A , MILLER E L , LONG D D E , et al. Effcient metadata management in large distributed file systems[C]// IEEE Symposium on Mass Storage Systems, April 7-10,2003, San Diego,USA. New Jersey: IEEE Press, 2003:290-298.
[41]	LIU Z , ZHOU X M . metadata management method based on directory path[J]. Journal of Software, 2007,18(2):236-245.
[42]	LI X Q , DONG B , XIAO L M , et al. Adaptive tradeoff in metadata-based small file optimizations for a cluster file system[J]. International Journal of Numerical Analysis and Modeling, 2012,9(2):289-303.
[43]	DEVULAPALLI A , WYCKO P . File creation strategies in a distributed metadata file system[C]// IEEE International Parallel and Distributed Processing Symposium(IEEE IPDPS 2007), March 26-30,2007, Rome,Italy. New Jersey: IEEE Press, 2007:1-10.
[3]	CHING A CHOUDHARY A , LIAO W K , et al. Effcient structured data access in parallel file systems[C]// The IEEE International Conference on Cluster Computing and the Grid, May 12-15,2003, Tokyo,Japan. New Jersey: IEEE Press, 2003:326-335.
[4]	THAKUR R GROPP W , LUSK E . Effcient structured data access in parallel file systems[C]// The 7th Symposium on the Frontiers of Massively Parallel Computation, February 26,1999, Annapolis,USA. New Jersey: IEEE Press, 1999:182-189.
[44]	YI L T , SHU J W , OU J X , et al. Cx:concurrent execution for the cross-server operations in a distributed file system[C]// 2012 IEEE International Conference on Cluster Computing, September 24-28,2012, Beijing,China. New Jersey: IEEE Press, 2012:99-107.
[45]	STAERELING R V H V , APPUSWAMY R , MOOLENBROEK D C V , et al. Efficient,modular metadata management with loris[C]// The 6th IEEE International Conference on Networking,Architecture and Storage, July 28-30,2011, Dalian,China. New Jersey: IEEE Press, 2011:278-287.
[5]	LIAO W K COLOMA K , CHOUDHARY A , et al. Collective caching:applicationaware client-side file caching[C]// The 14th International Symposium on High Performance Distributed Computing, October 24,2005, North Carolina,USA. New Jersey: IEEE Press, 2005:81-90.
[6]	CHEN Y ROTH P C . Collective prefetching for parallel I/O systems[C]// The 5th Petascale Data Storage Workshop(PDSW'10), November 15,2010, New Orleans,USA. New Jersey: IEEE Press, 2010:1-5.
[46]	STENDER J , KOLBECK B , HOGQVIST M . BabuDB:fast and effcient file system metadata storage[C]// 2010 International Workshop on Storage Network Architecture and Parallel I/Os, May 3,2010, Incline Village,USA. New Jersey: IEEE Press, 2010:51-58.
[47]	CHEN P M , PATTERSON D A . Maximizing performance in a striped disk array[J]. SIGARCH Computer Architecture News, 1990,18(2SI):322-331.
[48]	ROSS R B , CARNS P H , THAKUR R . PVFS:a parallel file system for linux clusters[C]// The 4th Annual Linux Showcase and Conference, October 10-14,2000, Atlanta,Georgia. Berkeley: USENIX Association Berkeley, 2000:391-430.
[49]	SCHEUERMANN P , WEIKUM G , ZABBACK P . Data partitioning and load balancing in parallel disk systems[J]. The VLDB Journal, 1998,7(1):48-66.
[50]	VASQUEZ H , LUDWIG T . Hint controlled distribution with parallel file systems[C]// Recent Advances in Parallel Virtual Machine and Message Passing Interface, September 18-21,2005, Sorrento,Italy. Berlin: Springer, 2005:110-118.
[51]	DONG B , LI X , XIAO L , et al. A new filespecific stripe size selection method for highly concurrent data access[C]// IEEE 13th International Conference on Grid Computing, September 20-23,2012, Beijing,China. New Jersey: IEEE Press, 2012:22-30.
[52]	LEE L . File assignment in parallel I/O systems with minimal variance of service time[J]. IEEE Transactions on Computers, 2000,49(2):127-140.
[53]	KUNKEL J M . Towards automatic load balancing of a parallel file system with subfile based migration[D]. Heidelberg:Heidelberg University, 2007.
[54]	SCHEUERMANN P , WEIKUM G , ZABBACK P . Data partitioning and load balancing in parallel disk systems[J]. The VLDB Journal, 1998,7(1):48-66.
[55]	SCHMUCK F B , HASKIN R . GPFS:a shared-disk file system for large computing clusters[C]// Conference on File and Storage Technologies, January 28-30,2002, Monterey,USA. Berkeley: USENIX Association Berkeley, 2002:19.
[56]	CARNS P H , III LIGON W B , ROSS R B , et al. PVFS:a parallel file system for Linux Clusters[C]// The 4th Annual Linux Showcase and Conference, October 10-14,2000, Atlanta,USA. Berkeley: USENIX Association Berkeley, 2000:391-430.
[57]	LIU W , WU M , OU X , et al. esign of an I/O balancing file system on Web server clusters[C]// The 2000 International Workshop on Parallel Processing, August 21-24,2000, Toronto,Canada. New Jersey: IEEE Press, 2000:119-126.
[58]	SCHROEDER B , GIBSON G A . A large-scale study of failures in highperformance computing systems[J]. International Conference on Dependable Systems ＆ Networks, 2010,7(4):337-351.
[59]	DONG B , LI X Q , WU Q M , et al. A dynamic and adaptive load balancing strategy for parallel file system with large-scale I/O servers[J]. Journal of Parallel and Distributed Computing, 2012,72(10):1254-1268.
[60]	DOWDY L W , FOSTER D V . Comparative models of the file assignment problem[J]. ACM Computing Surveys, 1982,14(2):287-313.
[61]	VERMA A , ANAND A . General store placement for response time minimization in parallel disks[J]. Journal of Parallel ＆Distributed Computing, 2007,67(12):1286-1300.
[62]	LEE L W , SCHEUERMANN P , VINGRALEK R . File assignment in parallel I/O systems with minimal variance of service time[J]. IEEE Transactions on Computers, 2000,49(2):127-140.
[63]	XIE T , SUN Y . A file assignment strategy independent of workload characteristic assumptions[J]. ACM Transactions on Storage, 2009,5(3):1-24.
[64]	DONG B , LI X Q , XIAO L M , et al. An optimal candidate selection model for self-acting load balancing of parallel file system[J]. International Journal of High Performance Computing and Networking, 2012,7(2):123-128.
[65]	CAMPELLO D , LOPEZ H , USECHE L , et al. Non-blocking writes to files[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:151-165.
[66]	BHAGWAT D , PATIL M , OSTROWSKI M , et al. A practical implementation of clustered fault tolerant write acceleration in a virtualized environment[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:287-300.
[67]	LI Q , SHI L , XUE C J , et al. Access characteristic guided read and write cost regulation for performance improvement on flash memory[C]// The 14th USENIX Conference on File and Storage Technologies, February 22-25,2016, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2016:125-132.
[68]	JANNEN W , YUAN J , ZHAN Y , et al. BetrFS:a right-optimized writeoptimized file system[C]// The 13th USENIX Conference on File and Storage Technologies, February 16-19,2015, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2015:301-315.
[69]	YUAN J , ZHAN Y , JANNEN W , et al. Optimizing every operation in a writeoptimized file system[C]// The 14th USENIX Conference on File and Storage Technologies, February 22-25,2016, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2016:1-14.
[70]	LEE S K , LIM K H , SONG H , et al. WORT:write optimal radix tree for persistent memory storage systems[C]// The 15th USENIX Conference on File and Storage Technologies, February 27 March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:257-270.
[71]	DOUGLIS F , DUGGAL A , SHILANE P , et al. The logic of physical garbage collection in deduplicating storage[C]// The 15th USENIX Conference on File and Storage Technologies, February 27-March 2,2017, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2017:29-44.
[72]	RUMBLE S M , KEJRIWAL A , OUSTERHOUT J K , et al. Log-structured memory for DRAM-based storage[C]// The 12th Usenix Conference on File and Storage Technologies, February 17-20,2014, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2014:1-16.
[73]	CULLY B , WIRES J , MEYER D T , et al. Strata:scalable high-performance storage on virtualized non-volatile memory[C]// The 12th USENIX Conference on File and Storage Technologies, February 17-20,2014, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2014:17-31.
[74]	CAULFIELD A M , DE A , COBURN J , et al. Moneta:a high-performance storage array architecture for nextgeneration,non-volatile memories[C]// The 43rd Annual IEEE/ACM International Symposium on Microarchitecture (2010), December 4-8,2010, Atlanta,USA. New Jersey: IEEE Press, 2010:385-395.
[75]	KIM H , SESHADRI S , DICKEY C L , et al. Evaluating phase change memory for enterprise storage systems:a study of caching and tiering approaches[J]. ACM Transactions on Storage, 2014,10(4):1-21.
[76]	DEMERS A , KESHAV S , SHENKER S . Analysis and simulation of a fair queueing algorithm[J]. ACM SIGCOMM Computer Communication Review, 1989,19(4):1-12.
[77]	GOYAL P , VIN H M , CHEN H . Starttime fair queueing:a scheduling algorithm for integrated services packet switching networks[J]. ACM SIGCOMM Computer Communication Review, 1996,26(4):157-168.
[78]	GULATI A , AHMAD I , WALDSPURGER C A . PARDA:proportional allocation of resources for distributed storage access[C]// The 7th USENIX Conference on File and Storage Technologies, February 24-27,2009, San Francisco,USA. New Jersey: Berkeley:USENIX Association, 2009:85-98.
[79]	GULATI A , KUMAR C , AHMAD I , et al. BASIL:automated IO load balancing across storage devices[C]// The 8th USENIX Conference on File and Storage Technologies, February 23-26,2010, San Jose,USA. Berkeley: USENIX Association Berkeley, 2010:169-182.
[80]	GULATI A , SHANMUGANATHAN G , ZHANG X , et al. Demand based hierarchical QoS using storage resource pools[C]// The 2012 USENIX Conference on Annual Technical Conference, June 13-15,2012, Boston,USA. Berkeley: USENIX Association Berkeley, 2012:1-13.
[81]	GHODSI A , ZAHARIA M , HINDMAN B , et al. Dominant resource fairness:fair allocation of multiple resource types[C]// The 8th USENIX Conference on Networked Systems Design and Implementation, March 30-April 1,2011, Boston,USA. Berkeley: USENIX Association Berkeley, 2011:323-336.
[82]	GHODSI A , SEKAR V , ZAHARIA M , et al. Multi-resource fair queueing for packet processing[J]. ACM SIGCOMM Computer Communication Review, 2012,42(4):1-12.
[83]	DOLEV D , FEITELSON D G , HALPERN J Y , et al. No justified complaints:on fair sharing of multiple resources[C]// The 3rd Innovations in Theoretical Computer Science Conference, January 8-10,2012, Cambridge,UK. New York: ACM Press, 2012:68-75.
[84]	GUTMAN A , NISAN N . Fair allocation without trade[C]//The 11th International Conference on Autonomous Agents and Multiagent Systems[C]// The 11th International Conference on Autonomous Agents and Multiagent Systems, June 4-8,2012, Valencia,Spain. New York: ACM Press, 2012:719-728.
[85]	PSOMAS C A , SCHWARTZ J . Beyond beyond dominant resource fairness:indivisible resource allocation in clusters[R]. Berkeley:Tech Report Berkeley, 2013.
[86]	PARKES D C , PROCACCIA A D , SHAH N . Beyond dominant resource fairness:extensions,limitations,and indivisibilities[J]. ACM Transactions on Economics and Computation, 2015,3(1):3.
[87]	WANG H , VARMAN P . Balancing fairness and effciency in tiered storage systems with bottleneck-aware allocation[C]// The 12th USENIX Conference on File and Storage Technologies, February 17-20,2014, Santa Clara,USA. Berkeley: USENIX Association Berkeley, 2014:229-242.
[88]	WANG W , LIANG B , LI B . Multi-resource fair allocation in heterogeneous cloud computing systems[J]. IEEE Transactions on Parallel and Distributed Systems, 2015,26(10):2822-2835.
[89]	HUO Z S , XIAO L M , ZHONG Q L , et al. Hybrid storage throughput allocation among multiple clients in heterogeneous data center[C]// IEEE 7th International Symposium on High Performance Computing and Communications, August 24-26,2015, New York,USA. New Jersey: IEEE Press, 2015:140-147.
[90]	VAKALI A . Evolutionary techniques for web caching[J]. Distributed ＆ Parallel Databases, 2002,11(1):93-116.
[91]	MADHYASTHA T M , REED D A . Learning to classify parallel input/output access patterns[J]. IEEE Transactions on Parallel ＆ Distributed Systems, 2002,13(8):802-813.
[92]	SETTLEMYER B W . A study of clientbased caching for parallel I/O[D]. Clemson:Clemson University, 2009.
[93]	VILAYANNUR M , SIVASUBRAMANIAM A , KANDEMIR M , et al. Discretionary caching for I/O on clusters[J]. Cluster Computing, 2006,9(1):29-44.
[94]	WANG X , MALIK T , BURNS R , et al. A workload-driven unit of cache replacement for mid-tier database caching[C]// The 12th International Conference on Database Systems for Advanced Applications(DASFAA'07), April 9-12,2007, Bangkok,Thailand. Berlin: Springer-Verlag, 2007:374-385.
[95]	SETTLEMYER B W . study of clientbased caching for parallel I/O[D]. Clemson:Clemson University, 2009.
[96]	VILAYANNUR M , SIVASUBRAMANIAM A , KANDEMIR M , et al. Discretionary caching for i/o on clusters[J]. Cluster Computing, 2006,9(1):29-44.
[97]	MADHYASTHA T M , REED D A . Learning to classify parallel input/output access patterns[J]. IEEE Transactions on Parallel ＆ Distributed Systems, 2002,13(8):802-813.
[98]	WACHS M , ABD-EL-MALEK M , THERESKA E , et al. Argon:performance insulation for shared storage servers[C]// The 5th USENIX Conference on File and Storage Technologies(FAST'07), February 13-16,2007, San Jose,USA. Berkeley: USENIX Association Berkeley, 2007:5.
[99]	VILAYANNUR M , NATH P , SIVASUBRAMANIAM A . Providing tunable consistency for a parallel file store[C]// The 3rd USENIX conference on File and Storage Technologies(FAST'07) (FAST'05), December 13-16,2005, San Francisco,USA. Berkeley: USENIX Association Berkeley, 2005:17-30.
[100]	FRASCA M , PRABHAKAR R , RAGHAVAN P , et al. Virtual I/O caching:dynamic storage cache management for concurrent workloads[C]// 2011 International Conference for High Performance Computing,Networking,Storage and Analysis, December 29,2011, Seatle,USA. New Jersey: IEEE Press, 2011:1-11.
[101]	LI M , VARKI E , BHATIA S , et al. Tap:table-based prefetching for Storage caches[C]// The 6th USENIX Conference on File and Storage Technologies, February 26-29,2008, San Jose,USA. Berkeley: USENIX Association Berkeley, 2008:1-16.
[7]	THAKUR R ROSS R , LUSK E , et al. Users guide for ROMIO:a high-performance,portable mpi-io implementation[R].[S.l.]: UNT Libraries Government Documents Department, 2004.
[102]	LI X Q , DONG B , XIAO L M , et al. Small files problem in parallel file system[C]// The 2011 International Conference on Network Computing and Information Security, May 14-15,2011, Guilin,China. New Jersey: IEEE Press, 2011:227-234.