一种面向OBS的可靠存储协议-RH-SCSI
|
摘要
随着计算机技术和网络技术的发展对存储系统提出了越来越高的要求,面向对象存储系统(Object Based storage System,OBS)是适应于下一代互联网环境的新型网络存储技术。OBS由计算节点、对象存储设备(Object-based Storage Device,OSD)、元数据服务器、存储网络四个部分组成。OBS以对象接口为数据访问方式,从而可充分利用数据对象属性来提高存储节点的自治性和主动性,而存储网络是对象存储系统的重要基础设施,其性能对存储系统的整体性能有重大的影响。
在网络存储协议HyperSCSI(Hyper Small Computer System Interface)的基础上,为面向对象存储系统设计了一种改进的网络存储协议—可靠加强型小型计算机系统接口协议(Reliable Hyper Small Computer System Interface,RH-SCSI)。HyperSCSI与网络存储协议iSCSI相比,HyperSCSI具有更高的性能。但HyperSCSI也有不足,其错误恢复机制不强,在提高性能的同时可靠性方面有所欠缺,同时在扩展性方面也有限制。
针对HyperSCSI的不足,设计的RH-SCSI协议在HyperSCSI层加入错误校验和恢复(Error Checking and Correcting,ECC)机制,提高了协议的可靠性。当接收端接收到的大数据包出现错误时,包头和包内数据块单元的头部出现的错误将被检测定位,然后被纠正;当接收端收到的小数据出现错误时将会采用重传机制来解决。这个机制可以提升存储网络的传输效率,减轻了网络的负担。在网络仿真平台NS-2上对两种协议进行了仿真,对仿真结果进行了分析。仿真结果表明RH-SCSI可以提高传输效率和可靠性,具有较好的错误恢复机制。
Storage systems have faced more and more challenges with the development of computer technique and networks, including higher I/O speed, larger storage capacity, security, expansibility and usability. The OBS(Object Based storage System) fits in with the next generation Internet. The OBS makes up of computer node, object-based storage device(OSD), metadata server and network. The storage network is the important factor of affecting the performance of OBS.
We present a RH-SCSI(Reliable HyperSCSI) protocol, which is designed based on HyperSCSI. It can establish high-performance storage system, because it was desighed for transmitting SCSI data and it makes use of existing Ethernet infrastructure, existing hardware part. It has shown to have more stronger functions initially than iSCSI. But HyperSCSI is insufficient in reliability.
To overcome the shortcoming of it, we present a RH-SCSI(Reliable HyperSCSI) protocol, which adds ECC(Error Checking and Correcting) mechanism to HyperSCSI layer. The strategy of RH-SCSI is that adding ECC mechanism to HyperSCSI layer. While the large size packet data received is error, error data could be checked and corrected. In case of little size data and command packets are error, RH-SCSI source re-transmits the error packets based on acknowle-dgement policy. These mechanism raises the throughput of the storage network, lightens the burden of the network. The performance of RH-SCSI is analyzed, and the simulation results on NS-2 show that the presented protocol is efficient.
在网络存储协议HyperSCSI(Hyper Small Computer System Interface)的基础上,为面向对象存储系统设计了一种改进的网络存储协议—可靠加强型小型计算机系统接口协议(Reliable Hyper Small Computer System Interface,RH-SCSI)。HyperSCSI与网络存储协议iSCSI相比,HyperSCSI具有更高的性能。但HyperSCSI也有不足,其错误恢复机制不强,在提高性能的同时可靠性方面有所欠缺,同时在扩展性方面也有限制。
针对HyperSCSI的不足,设计的RH-SCSI协议在HyperSCSI层加入错误校验和恢复(Error Checking and Correcting,ECC)机制,提高了协议的可靠性。当接收端接收到的大数据包出现错误时,包头和包内数据块单元的头部出现的错误将被检测定位,然后被纠正;当接收端收到的小数据出现错误时将会采用重传机制来解决。这个机制可以提升存储网络的传输效率,减轻了网络的负担。在网络仿真平台NS-2上对两种协议进行了仿真,对仿真结果进行了分析。仿真结果表明RH-SCSI可以提高传输效率和可靠性,具有较好的错误恢复机制。
Storage systems have faced more and more challenges with the development of computer technique and networks, including higher I/O speed, larger storage capacity, security, expansibility and usability. The OBS(Object Based storage System) fits in with the next generation Internet. The OBS makes up of computer node, object-based storage device(OSD), metadata server and network. The storage network is the important factor of affecting the performance of OBS.
We present a RH-SCSI(Reliable HyperSCSI) protocol, which is designed based on HyperSCSI. It can establish high-performance storage system, because it was desighed for transmitting SCSI data and it makes use of existing Ethernet infrastructure, existing hardware part. It has shown to have more stronger functions initially than iSCSI. But HyperSCSI is insufficient in reliability.
To overcome the shortcoming of it, we present a RH-SCSI(Reliable HyperSCSI) protocol, which adds ECC(Error Checking and Correcting) mechanism to HyperSCSI layer. The strategy of RH-SCSI is that adding ECC mechanism to HyperSCSI layer. While the large size packet data received is error, error data could be checked and corrected. In case of little size data and command packets are error, RH-SCSI source re-transmits the error packets based on acknowle-dgement policy. These mechanism raises the throughput of the storage network, lightens the burden of the network. The performance of RH-SCSI is analyzed, and the simulation results on NS-2 show that the presented protocol is efficient.
引文
[1] Wang F, Scott A, Brandt, et al. OBFS: A File System for Object-based Storage Devices. in: Harano G, ed. Proceedings of the 21st IEEE/12th NASA Goddard Conference on Mass Storage Systems and Technologies. MD: College Park, 2004. 283~300
[2] Gibson G A, Nagle D F, Courtright W, et al. NASD Scalable Storage Systems. in: Vincze M, Hager G, eds. Proceedings of USENIX 1999. Monterey, CA: Linux Workshop, 1999. 188~204
[3] Wang W, Yeo H, Zhu Y, et al. Design and development of Ethernet-based storage area network protocol. Computer Communications, 2003, 29(9): 1271~1283
[4] Telikepalli R, Drwjega T, Yan J. Storage area network extension solutions and their performance assessment. IEEE Communications Magazine, 2004, 10(4): 45~50
[5] Cai T, Ju S G, Zhao J J, et al. Self-certifying object-based network storage. in: Xiao N, ed. Proceedings of the 5th International Conference on Grid and Cooperative Computing. Hunan, China: IEEE Computer Society, 2006. 419~423
[6] Banerjee A, Drake L, Lang L, et al. GMPLS: an overview of signalling enhancements and recovery techniques. IEEE Communications Magazine, 2001, 17(4): 144~151
[7] Dahlin M, Wang R, Anderson T E, et al. Cooperative Caching: Using Remote Client Memory to Improve File System Performance. in: Jay L, ed. Proceedings of the First Symposium on Operating Systems Design and Implementation. Monterey, CA, United States: USENIX Association, 1994. 267~280
[8] Menon J, Pease D A, Rees R, et al. IBM Storage Tank-A heterogeneous scalable SAN file system. IBM Systems Journal, 2003, 42(2): 250~267
[9] T10 Work Group. SCSI Object-Based Storage Device Commands-2. T10/1355-D working draft, 2004
[10] Buyya R.高性能集群计算:结构与系统. (第一版).郑纬民,石威,汪东升等译.北京:电子工业出版社, 2001. 2~31
[11] Kai H. Advanced Computer Architecture Parallelism Scalability Programmability. McGraw-Hill Inc, 1993. 63~96
[12] Cuenca-Acuna F M, Nguyen T D. Cooperative Caching Middleware for Cluster- Based Servers. in: Anon, ed. The 10th IEEE international Symposium on High Performance Distributed Computing. San Francisco, USA: IEEE Computer Society Press, 2001. 303~314
[13] Voelker G, Anderson E, Kimbrel T, et al. Implementing Cooperative Prefetching and Caching in a Globally Managed Memory System. in: Leutenegger S, ed. Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems. Madison, Wisconsin: ACM, 1998. 33~43
[14] Gibson G A, Nagle D F, Amiri K, et al. File Server Scaling with Network-Attached Secure Disks. in: Leutenegger S, ed. Proceedings of the ACM International Conference on Measurement and Modeling of Computer System. New York, USA: ACM, 1997. 272~284
[15] Sarkar P, Hartman J H. Hint-based Cooperative Caching. ACM Transactions on Computer Systems, 2000, 18(4): 387~419
[16] Saker K. An Analysis of Object Storage Architecture. IEEE Computer, 2003, 2(3): 23~36
[17] Hospodor A D, Miller E L. Interconnection Architectures for Petabyte-Scale High-Performance Storage Systems. in: Harano G, ed. Proceedings of the 21st IEEE 12th NASA Goddard Conference on Mass Storage Systems and Technologies. MD: College Park, 2004. 273~281
[18]何飞跃.并行文件系统元数据管理研究: [硕士学位论文].武汉:华中科技大学图书馆, 2004. 35~46
[19] Cuneyt A, Sarit M. A Scalable Bandwith Guaranteed Distributed Continus Media File System Using Network Attached Autonomous Disks. IEEE Transactions on Multimedia, 2003, 5(1): 71~96
[20] Gibson G A, Nagle D F, Courtright II W, et al. NASD Scalable Storage Systems. in:Vincze M, Hager G, eds. Proceedings of USENIX 1999. Monterey, CA: Linux Workshop, 1999. 188~204
[21] Menon J, Pease D A, Rees R, et al. IBM Storage Tank-A heterogeneous scalable SAN file system. IBM Systems Journal, 2003, 42(2): 250~261
[22] Azagury A, Dreizin V, Factor M, et al. Towards an object store. in: Titsworth F. Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies. San Diego: IEEE, 2003. 165~176
[23] Terry J, Alice K, Yates R K. IBM通用并行文件系统的性能.齐丽红译.电子计算机, 2000, 25(10): 53~60
[24] Morris J H, Satyanarayanan M, Conner M H, et al. Andrew: A Distributed Personal Computing Environment. Communications of the ACM, 1986, 29(3): 184~201
[25] Corbett P F, Feitelso D G.. The Vesta Parallel File System. ACM Transactions on Computer Systems, 1996, 14(3): 225~264
[26] Brandt S A, Xue L, Miller E L, et al. Efficient Metadata Management in Large Distributed File Systems. in: Titsworth F, ed. Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies. San Diego, California: IEEE Computer Society, 2003. 290~298
[27]王琰.面向对象磁盘控制器软件系统研究与实现: [硕士学位论文].武汉:华中科技大学图书馆, 2006. 3~60
[28] Yan J, Zhu Y L, Xiong H. A Design of Metadata Server Cluster in Large Distributed Object-based Storage. in: Harano G, ed. Proceedings of 21st IEEE/12th NASA Goddard Conference on Mass Storage Systems and Technologies. MD: College Park, 2004. 13~16
[29]谈华芳,孙丽丽,侯紫峰.一种基于对象存储中的元数据组织管理方法.计算机工程, 2005, 31(20): 77~79
[30] Zhou Gongye, Chen Peng. RH-SCSI: A Reliable HyperSCSI Protocol for Networking Storage. NAS’2007,已录用.
[31] Yang Q, Henry, Zhang H. Integrating Web Prefetching and Caching Using Prediction Models. World Wide Web, 2001, 12(4): 299~321
[32] Milutinovic V, Valero M. Guest Editors Introduction-Cache Memory and RelatedProblems: Enhancing and Exploiting the Locality. IEEE Transactions on Compute- rs , February 1999, 48(2): 97~99
[33] Sherwood T, Perelman E, Calder B. Basic Block Distribution Analysis to Find Periodic Behavior and Simulation Points in Applications. in: Anon, ed. Proceedings of the International Conference on Parallel Architecture and Compilation Techniques. Washington, DC, USA: IEEE Computer Society Press, 2001. 3~14
[34] Thakur R, Gropp W, Lusk E. Data Sieving and Collective I/O in ROMIO. in: Anon, ed. The 7th Symposium on the Frontiers of Massively Parallel Computation. Washington, DC, USA: IEEE Computer Society Press, 1999. 182~189
[35] Markatos E P. On caching search engine query results. Computer Communications, 2001, 24(2): 137~143
[36] Zhou Y, Philbin J F, Li K. The Multi-Queue Replacement Algorithm for Second Level Buffer caches. in: Anderson E, Couch A, eds. Proceedings of the 2001 USENIX Annual Technical Conference. Boston, Massachusetts, USA. Berkeley: USENIX Association, 2001. 91~104
[37] Acharya A, Uysal M, Saltz J. Active disks: programming model, algorithms and evaluation. in: Anon, ed. Proceedings of the 8th Conference on Architectural Support for Programming Languages and Operating System. New York: IEEE, 1998. 81~91
[38] Riedel E, Faloutsos C, Gibson G, et al. Active disks for large-scale data processing. IEEE Computer, 2001, 34(6): 68~74
[39] Akinlar C, Mukherjee S. A scalable bandwidth guaranteed distributed continuous media file system using network attached autonomous disks. IEEE Transactions on Multimedia, 2003, 5(1): 71~96
[40] Qin L J, Feng D. Active storage framework for object-based storage device. in: Anon, ed. Proceedings of the 20th International Conference on Advanced Information Networking and Applications. Piscataway: IEEE, 2006. 97~101
[2] Gibson G A, Nagle D F, Courtright W, et al. NASD Scalable Storage Systems. in: Vincze M, Hager G, eds. Proceedings of USENIX 1999. Monterey, CA: Linux Workshop, 1999. 188~204
[3] Wang W, Yeo H, Zhu Y, et al. Design and development of Ethernet-based storage area network protocol. Computer Communications, 2003, 29(9): 1271~1283
[4] Telikepalli R, Drwjega T, Yan J. Storage area network extension solutions and their performance assessment. IEEE Communications Magazine, 2004, 10(4): 45~50
[5] Cai T, Ju S G, Zhao J J, et al. Self-certifying object-based network storage. in: Xiao N, ed. Proceedings of the 5th International Conference on Grid and Cooperative Computing. Hunan, China: IEEE Computer Society, 2006. 419~423
[6] Banerjee A, Drake L, Lang L, et al. GMPLS: an overview of signalling enhancements and recovery techniques. IEEE Communications Magazine, 2001, 17(4): 144~151
[7] Dahlin M, Wang R, Anderson T E, et al. Cooperative Caching: Using Remote Client Memory to Improve File System Performance. in: Jay L, ed. Proceedings of the First Symposium on Operating Systems Design and Implementation. Monterey, CA, United States: USENIX Association, 1994. 267~280
[8] Menon J, Pease D A, Rees R, et al. IBM Storage Tank-A heterogeneous scalable SAN file system. IBM Systems Journal, 2003, 42(2): 250~267
[9] T10 Work Group. SCSI Object-Based Storage Device Commands-2. T10/1355-D working draft, 2004
[10] Buyya R.高性能集群计算:结构与系统. (第一版).郑纬民,石威,汪东升等译.北京:电子工业出版社, 2001. 2~31
[11] Kai H. Advanced Computer Architecture Parallelism Scalability Programmability. McGraw-Hill Inc, 1993. 63~96
[12] Cuenca-Acuna F M, Nguyen T D. Cooperative Caching Middleware for Cluster- Based Servers. in: Anon, ed. The 10th IEEE international Symposium on High Performance Distributed Computing. San Francisco, USA: IEEE Computer Society Press, 2001. 303~314
[13] Voelker G, Anderson E, Kimbrel T, et al. Implementing Cooperative Prefetching and Caching in a Globally Managed Memory System. in: Leutenegger S, ed. Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems. Madison, Wisconsin: ACM, 1998. 33~43
[14] Gibson G A, Nagle D F, Amiri K, et al. File Server Scaling with Network-Attached Secure Disks. in: Leutenegger S, ed. Proceedings of the ACM International Conference on Measurement and Modeling of Computer System. New York, USA: ACM, 1997. 272~284
[15] Sarkar P, Hartman J H. Hint-based Cooperative Caching. ACM Transactions on Computer Systems, 2000, 18(4): 387~419
[16] Saker K. An Analysis of Object Storage Architecture. IEEE Computer, 2003, 2(3): 23~36
[17] Hospodor A D, Miller E L. Interconnection Architectures for Petabyte-Scale High-Performance Storage Systems. in: Harano G, ed. Proceedings of the 21st IEEE 12th NASA Goddard Conference on Mass Storage Systems and Technologies. MD: College Park, 2004. 273~281
[18]何飞跃.并行文件系统元数据管理研究: [硕士学位论文].武汉:华中科技大学图书馆, 2004. 35~46
[19] Cuneyt A, Sarit M. A Scalable Bandwith Guaranteed Distributed Continus Media File System Using Network Attached Autonomous Disks. IEEE Transactions on Multimedia, 2003, 5(1): 71~96
[20] Gibson G A, Nagle D F, Courtright II W, et al. NASD Scalable Storage Systems. in:Vincze M, Hager G, eds. Proceedings of USENIX 1999. Monterey, CA: Linux Workshop, 1999. 188~204
[21] Menon J, Pease D A, Rees R, et al. IBM Storage Tank-A heterogeneous scalable SAN file system. IBM Systems Journal, 2003, 42(2): 250~261
[22] Azagury A, Dreizin V, Factor M, et al. Towards an object store. in: Titsworth F. Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies. San Diego: IEEE, 2003. 165~176
[23] Terry J, Alice K, Yates R K. IBM通用并行文件系统的性能.齐丽红译.电子计算机, 2000, 25(10): 53~60
[24] Morris J H, Satyanarayanan M, Conner M H, et al. Andrew: A Distributed Personal Computing Environment. Communications of the ACM, 1986, 29(3): 184~201
[25] Corbett P F, Feitelso D G.. The Vesta Parallel File System. ACM Transactions on Computer Systems, 1996, 14(3): 225~264
[26] Brandt S A, Xue L, Miller E L, et al. Efficient Metadata Management in Large Distributed File Systems. in: Titsworth F, ed. Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies. San Diego, California: IEEE Computer Society, 2003. 290~298
[27]王琰.面向对象磁盘控制器软件系统研究与实现: [硕士学位论文].武汉:华中科技大学图书馆, 2006. 3~60
[28] Yan J, Zhu Y L, Xiong H. A Design of Metadata Server Cluster in Large Distributed Object-based Storage. in: Harano G, ed. Proceedings of 21st IEEE/12th NASA Goddard Conference on Mass Storage Systems and Technologies. MD: College Park, 2004. 13~16
[29]谈华芳,孙丽丽,侯紫峰.一种基于对象存储中的元数据组织管理方法.计算机工程, 2005, 31(20): 77~79
[30] Zhou Gongye, Chen Peng. RH-SCSI: A Reliable HyperSCSI Protocol for Networking Storage. NAS’2007,已录用.
[31] Yang Q, Henry, Zhang H. Integrating Web Prefetching and Caching Using Prediction Models. World Wide Web, 2001, 12(4): 299~321
[32] Milutinovic V, Valero M. Guest Editors Introduction-Cache Memory and RelatedProblems: Enhancing and Exploiting the Locality. IEEE Transactions on Compute- rs , February 1999, 48(2): 97~99
[33] Sherwood T, Perelman E, Calder B. Basic Block Distribution Analysis to Find Periodic Behavior and Simulation Points in Applications. in: Anon, ed. Proceedings of the International Conference on Parallel Architecture and Compilation Techniques. Washington, DC, USA: IEEE Computer Society Press, 2001. 3~14
[34] Thakur R, Gropp W, Lusk E. Data Sieving and Collective I/O in ROMIO. in: Anon, ed. The 7th Symposium on the Frontiers of Massively Parallel Computation. Washington, DC, USA: IEEE Computer Society Press, 1999. 182~189
[35] Markatos E P. On caching search engine query results. Computer Communications, 2001, 24(2): 137~143
[36] Zhou Y, Philbin J F, Li K. The Multi-Queue Replacement Algorithm for Second Level Buffer caches. in: Anderson E, Couch A, eds. Proceedings of the 2001 USENIX Annual Technical Conference. Boston, Massachusetts, USA. Berkeley: USENIX Association, 2001. 91~104
[37] Acharya A, Uysal M, Saltz J. Active disks: programming model, algorithms and evaluation. in: Anon, ed. Proceedings of the 8th Conference on Architectural Support for Programming Languages and Operating System. New York: IEEE, 1998. 81~91
[38] Riedel E, Faloutsos C, Gibson G, et al. Active disks for large-scale data processing. IEEE Computer, 2001, 34(6): 68~74
[39] Akinlar C, Mukherjee S. A scalable bandwidth guaranteed distributed continuous media file system using network attached autonomous disks. IEEE Transactions on Multimedia, 2003, 5(1): 71~96
[40] Qin L J, Feng D. Active storage framework for object-based storage device. in: Anon, ed. Proceedings of the 20th International Conference on Advanced Information Networking and Applications. Piscataway: IEEE, 2006. 97~101
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |