广域网存储服务中数据传输中间件的研究与实现
|
摘要
随着网络技术的迅速发展,通过网络进行传输的信息量正呈指数级增长,网络存储系统的研究已经成为计算机领域的一个研究热点。下一代互联网时代的到来对网络存储提出了新的要求,那就是海量结点,超大容量,服务式存储,高可靠性,高可用性和高性能。
广域网存储服务USTOR(Ubiquitous Storage)基于IPv6在广域网范围构建,支持PB级的总存储容量和至少256个分布式存储节点。系统分为存储管理、元数据和数据传输中间件三个部分。存储管理建立在存储管理主动规范协议之上,对各种异构的存储设备进行统一管理,并收集存储节点的设备和网络信息提交给元数据管理部分;相关信息被元数据模块处理成为全局视图的变量,交给上层的动态负载平衡模块,该模块协调各个存储节点之间的负载平衡和实现用户存储空间的分配与调度;数据传输中间件为用户提供存储服务,主要是数据管理和数据快速、安全的传输。
作为USTOR的重要组成部分,数据传输中间件在开放、动态的IPv6网络环境下实现数据的快速存取和安全可靠的传输。数据传输中间件主要包括三个部分:数据传输协议,安全机制和移动Cache机制。数据传输协议实现了信息通信和数据传输;安全机制包括三方身份认证和安全传输机制,前者是对用户的身份进行检验,而后者是为了防止重要数据在传输过程中被窃取和篡改而设置的安全措施;移动Cache机制建立在协议之上,是对传输性能的优化,可以改善传输中间件的传输服务质量。
最后在实验室搭建的广域网存储服务USTOR平台上进行了测试。测试结果表明,数据传输中间件和传统的传输协议(如FTP)相比,在传输性能上有着一定的优势,使用移动Cache机制进行传输优化后,传输性能得到很大提高,传输的安全性在安全机制的保障下得到进一步提高。
Along with the rapid development of network tech, the information transferred by network is increasing at exponent level. The research on network storage system is becoming a new hotspot in computer architecture field. Also, the next generation of the Internet put forward the new requirement of network storage that is massive in node and capacity, storage service, high reliability, convenient usability and excellent performance.
The Storage Service USTOR (Ubiquitous Storage) for the Next Generation Internet is set up in IPv6 WAN, which supports the total storage capacity of PB class, and supports at least 256 distributed nodes. The system is divided into such three parts as the storage management, metadata management and the data transfer middleware. The storage management component based on the SMI-S (Storage Management Interface Specification) protocol is responsible for the unifying management of various storage equipments, and collecting from each storage note of the related data to the metadata management component. Related data is proceed by the metadata management component and then handed to the upper dynamic load-balancing component, which will moderate the load circumstances among storage nodes, and then hand over related data in user storage space to the data transfer middleware. The data transfer middleware provides storage services to customers, mainly for data management and fast, convenient and safe data delivery.
As an important part of the USTOR, data transfer middleware, in the open and dynamic IPv6 environment, implements rapid and long-range data access, meanwhile it transfers data with great safety and reliability, and assures the storage service. The storage middleware consists of three parts: the transfer protocol module, the mobile cache module and the security module. The transfer protocol module implements file management as well as file reading and writing. The security module is made up of three-side authentication and secure data transfer mechanism, among which the former one is to check the identity of users while the later one is a security method to protect the important data from being stolen or modified when it is on its way. The mobile cache module, which is the representation of high-quality service and timely, is built on this protocol, and it is an optimization of the performance of data transfer.
The testing results of the storage system platform built in the lab shows that the data transfer middleware has some advantage over traditional transfer protocol (e.g. FTP) on the transfer rate. With the optimization of cache, the performance of data transfer has been greatly enhanced and its security is ensured by the security module.
广域网存储服务USTOR(Ubiquitous Storage)基于IPv6在广域网范围构建,支持PB级的总存储容量和至少256个分布式存储节点。系统分为存储管理、元数据和数据传输中间件三个部分。存储管理建立在存储管理主动规范协议之上,对各种异构的存储设备进行统一管理,并收集存储节点的设备和网络信息提交给元数据管理部分;相关信息被元数据模块处理成为全局视图的变量,交给上层的动态负载平衡模块,该模块协调各个存储节点之间的负载平衡和实现用户存储空间的分配与调度;数据传输中间件为用户提供存储服务,主要是数据管理和数据快速、安全的传输。
作为USTOR的重要组成部分,数据传输中间件在开放、动态的IPv6网络环境下实现数据的快速存取和安全可靠的传输。数据传输中间件主要包括三个部分:数据传输协议,安全机制和移动Cache机制。数据传输协议实现了信息通信和数据传输;安全机制包括三方身份认证和安全传输机制,前者是对用户的身份进行检验,而后者是为了防止重要数据在传输过程中被窃取和篡改而设置的安全措施;移动Cache机制建立在协议之上,是对传输性能的优化,可以改善传输中间件的传输服务质量。
最后在实验室搭建的广域网存储服务USTOR平台上进行了测试。测试结果表明,数据传输中间件和传统的传输协议(如FTP)相比,在传输性能上有着一定的优势,使用移动Cache机制进行传输优化后,传输性能得到很大提高,传输的安全性在安全机制的保障下得到进一步提高。
Along with the rapid development of network tech, the information transferred by network is increasing at exponent level. The research on network storage system is becoming a new hotspot in computer architecture field. Also, the next generation of the Internet put forward the new requirement of network storage that is massive in node and capacity, storage service, high reliability, convenient usability and excellent performance.
The Storage Service USTOR (Ubiquitous Storage) for the Next Generation Internet is set up in IPv6 WAN, which supports the total storage capacity of PB class, and supports at least 256 distributed nodes. The system is divided into such three parts as the storage management, metadata management and the data transfer middleware. The storage management component based on the SMI-S (Storage Management Interface Specification) protocol is responsible for the unifying management of various storage equipments, and collecting from each storage note of the related data to the metadata management component. Related data is proceed by the metadata management component and then handed to the upper dynamic load-balancing component, which will moderate the load circumstances among storage nodes, and then hand over related data in user storage space to the data transfer middleware. The data transfer middleware provides storage services to customers, mainly for data management and fast, convenient and safe data delivery.
As an important part of the USTOR, data transfer middleware, in the open and dynamic IPv6 environment, implements rapid and long-range data access, meanwhile it transfers data with great safety and reliability, and assures the storage service. The storage middleware consists of three parts: the transfer protocol module, the mobile cache module and the security module. The transfer protocol module implements file management as well as file reading and writing. The security module is made up of three-side authentication and secure data transfer mechanism, among which the former one is to check the identity of users while the later one is a security method to protect the important data from being stolen or modified when it is on its way. The mobile cache module, which is the representation of high-quality service and timely, is built on this protocol, and it is an optimization of the performance of data transfer.
The testing results of the storage system platform built in the lab shows that the data transfer middleware has some advantage over traditional transfer protocol (e.g. FTP) on the transfer rate. With the optimization of cache, the performance of data transfer has been greatly enhanced and its security is ensured by the security module.
引文
[1] Milojicic D.. Microsoft's Jim Gray on computing's breakthroughs, lessons, and future. Distributed Systems Online, IEEE. 2004, 5(1). 4.1~4.8
[2] Qin Xin, EL Miller, T Schwarz, et al. Reliability Mechanisms for Very Large Storage Systems. In: The 20th IEEE/11th NASA Goddard conference on Mass Storage Systems and Technologies (MSST’03). USA: IEEE Computer Society, 2003. 146~156
[3] Atul Adya, William J. Bolosky, Miguel Castro, et al. FARSITE: Federated, Available, and Reliable Storage for an Incompletely Trusted Environment. 5th Symposium on Operating Systems Design and Implementation (OSDI2002). Boston, MA, 2002. 118～201
[4] A. Rowstron, P. Druschel. Storage management and caching in PAST, a large-scale, persistent peer-to-peer storage utility. ACM SIGOPS Operating Systems Review, 2001, 35(5). 188~201
[5]张江陵,冯丹.海量信息存储.(第一版).北京:科学出版社.2003.1～5
[6] Garth A. Gibson, Rodney Van Meter. Network Attached Storage Architecture. Communications of the ACM, 2000, 43(11). 37-45
[7] Garth A. Gibson, David F. Nagle, Khalil Amiri, et al. Filesystems for Network-attached Secure Disks. Technical Report CMU-CS, 1997. 1~18
[8] David A. Patterson. Massive Parallelism And Massive Storage Trends and Predictions For 1995 to 2000. Second International Conference on Parallel and Distributed Information System. 1993. 197～205
[9] Daniel Duffy, Nicko Acks, Vaughn Noga, et al. Beyond the Storage Area Networking: Data Intensive Computing in a Distributed Environment. In: Proceeding of the 22nd IEEE/13th NASA Goddard Conference on Mass Storage System Technologies (MSST’05). American: IEEE Computer Society, 2005. 232~236
[10] Clark T. IP SANs: A Guide to iSCSI, iFCP, and FCIP Protocols for Storage AreaNetwork [M]. Boston: Addison -Wesley, 2002. 47~86
[11]郑纬民,舒继武.下一代分布式智能网络存储系统的发展趋势.电信世界,2004, 17(8).16～23
[12] Rhea S., Wells C., Eaton P., et al. Maintenance-free global data storage. Internet Computing, IEEE. USA: IEEE Computer Society, 2001. 40～49
[13] John Kubiatowicz, David Bindel, Yan Chen, et al. Oceanstore: An architecture for global-scale persistent store. ACM SIGOPS Operating Systems Review, 2000, 34(5). 190～201
[14] Utard G.; Vernois A.. Data durability in peer to peer storage systems. IEEE International Symposium on. USA: IEEE Computer Society, 2004. 90～97
[15] Randriamaro C., Soyez O., Utard G., et al. Data distribution in a peer to peer storage system. IEEE International Symposium on. USA: IEEE Computer Society, 2005. 284～291
[16] Micah Beck, Jean-Patrick Gelas, Dustin Parr, et al. LoDN: Logistical Distribution Network. Workshop on Advanced Collaborative Environments, Nice, France, 2004. 88～94
[17] James S. Plank, Micah Beck, Wael R. Elwasif, et al. The Internet Backplane Protocol: Storage in the Network. presented at NetStore99: The Network Storage Symposium, Seattle, WA, 1999. 242～246
[18] Alessandro Bassi, Micah Beck, James S. Plank, et al. Internet Backplane Protocol: API 1.0. CS-01-455, University of Tennessee Departement of Computer Science, 2001. 324～332
[19] Alessandro Bassi, Micah Beck, Terry Moore, et al. The Logistical Backbone: Scalable Infrastructure for Global Data Grids. Asian Computing Science Conference 2002, Hanoi, Vietnam, 2002. 158～166
[20] Alessandro Bassi, Micah Beck, Laurent Lefevre, et al. Logistical Storage in Active Networking: a promising framework for network services. 3rd International Conference on Internet Computing (IC), USA, 2002. 204～210
[21] M. Beck, T. Moore, J. Plank, et al. Logistical Networking: Sharing More Than the Wires. The Kluwer International Series in Engineering and Computer Science. Boston: Kluwer Academic Publishers, 2000. 112～118
[22] Weimin Zheng, Jinfeng Hu, Ming Li. Granary: Architecture of Object Oriented Internet Storage Service. IEEE International Conference on E-Commerce Technology for Dynamic E-Business (CEC-EAST 2004, invited paper). USA: IEEE Computer Society, 2004. 294～297
[23]胡进锋.基于对等结构的广域网分布式存储系统研究:博士学位论文.北京:清华大学图书馆.2005.
[24] R. J. Honicky, E. L. Miller. A fast algorithm for online placement and reorganization of replicated data. Proceedings of the 17th International Parallel and Distributed Processing Symposium. Nice, France, 2003. 57~68
[25]韦华.存储管理的统一之路——SNIA加快推行的SMI计划.微电脑世界,2003,(16).14～20
[26] ZHANG Hui-cheng, XIE Xiang-hui. Research on Storage Management Initiative Specification and Its Application. High Performance Computing Technology. 2006 (4). 5～9
[27] Ke Zhou, Ze Deng. A Domain-Based Storage Management Framework for the WAN. Networking, Architecture, and Storages, 2006. NAS'06. International Workshop on. 2006. 63～64
[28] Jong-Geun Park, Chang-Won Ahn, Hee-Nam Cho, et al. A method for representing and transporting CIM operations using binary XML in the WBEM architecture. Advanced Communication Technology, 2006. ICACT 2006. The 8th International Conference. 2006. 4～5
[29] Keslassy I., Cheng-Shang Chang, McKeown N., et al. Optimal load-balancing. 24th Annual Joint Conference of the IEEE Computer and Communications Societies. USA: IEEE Computer Society, 2005. 1712～1722
[30] Jaiganesh Balasubramanian, Schmidt D. C., Dowdy L., et al. Evaluating theperformance of middleware load balancing strategies. Enterprise Distributed Object Computing Conference. 2004. 135～146
[31]苏康.一种基于消息中间件的数据传输平台设计方法.科学技术与工程,2006,6(7).882～885
[32]邱岩,王卫兵.基于消息中间件的数据传输技术.现代电子技术,2004,(15). 44～45
[33] Smith C.L., Robinson M.. The understanding of security technology and its applications. Proceedings. IEEE 33rd Annual 1999 International Carnahan Conference on. USA: IEEE Computer Society, 1999. 26～37
[34] Thapliyal H., Kamala R.V., Srinivas M.. RSA encryption/decryption in wireless networks using an efficient high speed multiplier. IEEE 33rd Annual 1999 International Carnahan Conference on. USA: IEEE Computer Society, 1999. 26～37
[35] Apostolopoulos G., Peris V., Saha D.. Transport layer security: how much does it really cost? Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. USA: IEEE Computer Society 1999. 717～725
[36] Tewari R., Vin H. M., Dan A., et al. Resource based caching for web servers. In: Proceedings of MMCN. USA, 1998. 191～205
[37] Otoo E., Shoshani A.. Accurate modeling of cache replacement policies in a data grid. In : Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies, USA: IEEE Computer Society, 2003. 10～19
[38] Cohen E. , Kaplan H. . Caching documents with variable sizes and fetching costs : An LP-based approach. Algorithmica, 2001, 41(4). 41～53
[39] Wang J. A survey of web caching schemes for the Internet [J]. Computer Communication Review, 1999, 29(5). 36～46
[40] Gwetzman, Seltzer M.. World Wide Web cache consistency. Proceedings of the USENIX Technical Conference. USA: 1996. 141～152
[41]冯丹,陈小庆,李洁琼.广域网存储系统中自适应缓存一致性的研究.微处理机(已录用).
[42]李军锋,姜昱明.代理服务器上的自适应存储缓冲技术[J].计算机工程与应用,2003,(11).165～166
[43] Burr W. E.. Cryptographic hash standards: where do we go from here? Security and Privacy Magazine, IEEE. 2006, 4(2). 88～91
[2] Qin Xin, EL Miller, T Schwarz, et al. Reliability Mechanisms for Very Large Storage Systems. In: The 20th IEEE/11th NASA Goddard conference on Mass Storage Systems and Technologies (MSST’03). USA: IEEE Computer Society, 2003. 146~156
[3] Atul Adya, William J. Bolosky, Miguel Castro, et al. FARSITE: Federated, Available, and Reliable Storage for an Incompletely Trusted Environment. 5th Symposium on Operating Systems Design and Implementation (OSDI2002). Boston, MA, 2002. 118～201
[4] A. Rowstron, P. Druschel. Storage management and caching in PAST, a large-scale, persistent peer-to-peer storage utility. ACM SIGOPS Operating Systems Review, 2001, 35(5). 188~201
[5]张江陵,冯丹.海量信息存储.(第一版).北京:科学出版社.2003.1～5
[6] Garth A. Gibson, Rodney Van Meter. Network Attached Storage Architecture. Communications of the ACM, 2000, 43(11). 37-45
[7] Garth A. Gibson, David F. Nagle, Khalil Amiri, et al. Filesystems for Network-attached Secure Disks. Technical Report CMU-CS, 1997. 1~18
[8] David A. Patterson. Massive Parallelism And Massive Storage Trends and Predictions For 1995 to 2000. Second International Conference on Parallel and Distributed Information System. 1993. 197～205
[9] Daniel Duffy, Nicko Acks, Vaughn Noga, et al. Beyond the Storage Area Networking: Data Intensive Computing in a Distributed Environment. In: Proceeding of the 22nd IEEE/13th NASA Goddard Conference on Mass Storage System Technologies (MSST’05). American: IEEE Computer Society, 2005. 232~236
[10] Clark T. IP SANs: A Guide to iSCSI, iFCP, and FCIP Protocols for Storage AreaNetwork [M]. Boston: Addison -Wesley, 2002. 47~86
[11]郑纬民,舒继武.下一代分布式智能网络存储系统的发展趋势.电信世界,2004, 17(8).16～23
[12] Rhea S., Wells C., Eaton P., et al. Maintenance-free global data storage. Internet Computing, IEEE. USA: IEEE Computer Society, 2001. 40～49
[13] John Kubiatowicz, David Bindel, Yan Chen, et al. Oceanstore: An architecture for global-scale persistent store. ACM SIGOPS Operating Systems Review, 2000, 34(5). 190～201
[14] Utard G.; Vernois A.. Data durability in peer to peer storage systems. IEEE International Symposium on. USA: IEEE Computer Society, 2004. 90～97
[15] Randriamaro C., Soyez O., Utard G., et al. Data distribution in a peer to peer storage system. IEEE International Symposium on. USA: IEEE Computer Society, 2005. 284～291
[16] Micah Beck, Jean-Patrick Gelas, Dustin Parr, et al. LoDN: Logistical Distribution Network. Workshop on Advanced Collaborative Environments, Nice, France, 2004. 88～94
[17] James S. Plank, Micah Beck, Wael R. Elwasif, et al. The Internet Backplane Protocol: Storage in the Network. presented at NetStore99: The Network Storage Symposium, Seattle, WA, 1999. 242～246
[18] Alessandro Bassi, Micah Beck, James S. Plank, et al. Internet Backplane Protocol: API 1.0. CS-01-455, University of Tennessee Departement of Computer Science, 2001. 324～332
[19] Alessandro Bassi, Micah Beck, Terry Moore, et al. The Logistical Backbone: Scalable Infrastructure for Global Data Grids. Asian Computing Science Conference 2002, Hanoi, Vietnam, 2002. 158～166
[20] Alessandro Bassi, Micah Beck, Laurent Lefevre, et al. Logistical Storage in Active Networking: a promising framework for network services. 3rd International Conference on Internet Computing (IC), USA, 2002. 204～210
[21] M. Beck, T. Moore, J. Plank, et al. Logistical Networking: Sharing More Than the Wires. The Kluwer International Series in Engineering and Computer Science. Boston: Kluwer Academic Publishers, 2000. 112～118
[22] Weimin Zheng, Jinfeng Hu, Ming Li. Granary: Architecture of Object Oriented Internet Storage Service. IEEE International Conference on E-Commerce Technology for Dynamic E-Business (CEC-EAST 2004, invited paper). USA: IEEE Computer Society, 2004. 294～297
[23]胡进锋.基于对等结构的广域网分布式存储系统研究:博士学位论文.北京:清华大学图书馆.2005.
[24] R. J. Honicky, E. L. Miller. A fast algorithm for online placement and reorganization of replicated data. Proceedings of the 17th International Parallel and Distributed Processing Symposium. Nice, France, 2003. 57~68
[25]韦华.存储管理的统一之路——SNIA加快推行的SMI计划.微电脑世界,2003,(16).14～20
[26] ZHANG Hui-cheng, XIE Xiang-hui. Research on Storage Management Initiative Specification and Its Application. High Performance Computing Technology. 2006 (4). 5～9
[27] Ke Zhou, Ze Deng. A Domain-Based Storage Management Framework for the WAN. Networking, Architecture, and Storages, 2006. NAS'06. International Workshop on. 2006. 63～64
[28] Jong-Geun Park, Chang-Won Ahn, Hee-Nam Cho, et al. A method for representing and transporting CIM operations using binary XML in the WBEM architecture. Advanced Communication Technology, 2006. ICACT 2006. The 8th International Conference. 2006. 4～5
[29] Keslassy I., Cheng-Shang Chang, McKeown N., et al. Optimal load-balancing. 24th Annual Joint Conference of the IEEE Computer and Communications Societies. USA: IEEE Computer Society, 2005. 1712～1722
[30] Jaiganesh Balasubramanian, Schmidt D. C., Dowdy L., et al. Evaluating theperformance of middleware load balancing strategies. Enterprise Distributed Object Computing Conference. 2004. 135～146
[31]苏康.一种基于消息中间件的数据传输平台设计方法.科学技术与工程,2006,6(7).882～885
[32]邱岩,王卫兵.基于消息中间件的数据传输技术.现代电子技术,2004,(15). 44～45
[33] Smith C.L., Robinson M.. The understanding of security technology and its applications. Proceedings. IEEE 33rd Annual 1999 International Carnahan Conference on. USA: IEEE Computer Society, 1999. 26～37
[34] Thapliyal H., Kamala R.V., Srinivas M.. RSA encryption/decryption in wireless networks using an efficient high speed multiplier. IEEE 33rd Annual 1999 International Carnahan Conference on. USA: IEEE Computer Society, 1999. 26～37
[35] Apostolopoulos G., Peris V., Saha D.. Transport layer security: how much does it really cost? Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. USA: IEEE Computer Society 1999. 717～725
[36] Tewari R., Vin H. M., Dan A., et al. Resource based caching for web servers. In: Proceedings of MMCN. USA, 1998. 191～205
[37] Otoo E., Shoshani A.. Accurate modeling of cache replacement policies in a data grid. In : Proceedings of the 20th IEEE/11th NASA Goddard Conference on Mass Storage Systems and Technologies, USA: IEEE Computer Society, 2003. 10～19
[38] Cohen E. , Kaplan H. . Caching documents with variable sizes and fetching costs : An LP-based approach. Algorithmica, 2001, 41(4). 41～53
[39] Wang J. A survey of web caching schemes for the Internet [J]. Computer Communication Review, 1999, 29(5). 36～46
[40] Gwetzman, Seltzer M.. World Wide Web cache consistency. Proceedings of the USENIX Technical Conference. USA: 1996. 141～152
[41]冯丹,陈小庆,李洁琼.广域网存储系统中自适应缓存一致性的研究.微处理机(已录用).
[42]李军锋,姜昱明.代理服务器上的自适应存储缓冲技术[J].计算机工程与应用,2003,(11).165～166
[43] Burr W. E.. Cryptographic hash standards: where do we go from here? Security and Privacy Magazine, IEEE. 2006, 4(2). 88～91