WO2006097037A1 - Systeme tampon de stockage en reseau expansible haut debit - Google Patents

Systeme tampon de stockage en reseau expansible haut debit Download PDF

Info

Publication number
WO2006097037A1
WO2006097037A1 PCT/CN2006/000384 CN2006000384W WO2006097037A1 WO 2006097037 A1 WO2006097037 A1 WO 2006097037A1 CN 2006000384 W CN2006000384 W CN 2006000384W WO 2006097037 A1 WO2006097037 A1 WO 2006097037A1
Authority
WO
WIPO (PCT)
Prior art keywords
storage network
unit
cache
high speed
speed storage
Prior art date
Application number
PCT/CN2006/000384
Other languages
English (en)
French (fr)
Inventor
Yaolong Zhu
Hui Xiong
Jie Yan
Original Assignee
Zhang, Jinkui
Zhou, Feng
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhang, Jinkui, Zhou, Feng filed Critical Zhang, Jinkui
Priority to US11/886,435 priority Critical patent/US8032610B2/en
Publication of WO2006097037A1 publication Critical patent/WO2006097037A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
    • G06F3/0656—Data buffering arrangements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0813—Multiuser, multiprocessor or multiprocessing cache systems with a network or matrix configuration
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0866—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches for peripheral storage systems, e.g. disk cache
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/061—Improving I/O performance
    • G06F3/0613—Improving I/O performance in relation to throughput
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/31—Providing disk cache in a specific location of a storage system
    • G06F2212/314—In storage network, e.g. network attached cache

Definitions

  • the present invention relates to an expandable high speed storage network cache system for use in a storage network, managed by a storage network manager.
  • BACKGROUND OF THE INVENTION High-capacity and centralized storage networks are an important trend in the development of storage technologies.
  • Current storage network multi-purpose disk array devices are used as general-purpose storage systems to solve the problem of storage capacity and transmission bandwidth.
  • the seek time of the head takes about 10 milliseconds, and the response rate of a hard disk drive to random requests is only about 100 requests per second. Even if network storage uses disk arrays, it is difficult to break the 1000 requests per second by allocating requests to individual disks.
  • the present invention provides an expandable high-speed storage network cache system that can break the existing limit of request response and realize high-speed response server input and output requests.
  • an expandable high-speed storage network cache system comprising: a high-speed storage network cache system composed of at least one high-speed storage network cache module connected to the storage network, by at least one storage a storage network manager composed of a network management module, the high-speed storage network cache module is composed of a hardware platform unit and a software platform unit, the hardware platform unit adopts an embedded system platform or a server platform, and the high-speed storage network cache system is stored by Network manager management for shared use by servers and storage systems connected to the storage network.
  • the high-speed storage network cache module is composed of a hardware platform unit and a software platform unit, and the hardware platform unit directly utilizes an embedded system platform or a server platform.
  • the hardware resources when coupled with the corresponding software platform unit, make the high-speed storage network cache system a network storage device that can be shared by all servers and storage systems in the storage network under the management of the storage network manager.
  • the network device is a virtual cache system, which utilizes the memory resources or extended memory resources of the embedded system platform or the server platform, instead of the hard disk storage device. Therefore, the present invention greatly breaks the bottleneck problem of slow seek of the hard disk drive. , achieves a high-speed response to the server's input and output requests.
  • the hardware platform unit includes a central processing unit, an interface hardware unit, a cache hardware unit, and a data guarantee unit.
  • the cache hardware unit is operated by a cache software unit by managing internal memory or extended memory of the system. Memory virtualization becomes the hard disk use; the interface hardware unit is set so that the high-speed storage network cache system can be connected to the storage network server and the hard disk storage system recognizes it through the management of the storage network manager and is used as a normal cache memory;
  • the central processor is set up to have the service function of reading and writing other memories. Because the high-speed storage network cache system is connected to the disk storage system, it is a "transparent" storage device, and the final data is transferred to the disk storage system.
  • the data transfer unit of the software platform unit adopts a cache optimization module to implement a certain optimization algorithm and reduce the occupation of the storage network bandwidth.
  • the network storage manager manages the entire storage network system, detects storage module (including high-speed storage network cache module and storage system) expansion; and controls the flow of data, requests for random data are sent to the high-speed storage network cache system, and continuous data The request is sent directly to the disk storage system.
  • the setting of the above technical solution can greatly improve the response rate of the random request.
  • FIG. 1 is a schematic structural diagram of a cache system that cannot implement storage system sharing.
  • Figure 2 is a schematic view showing the structure of the system of the present invention.
  • FIG. 3 is a structural block diagram of a high speed storage network cache module of the present invention.
  • FIG. 4 is a schematic diagram of a system of the present invention applied to a storage network.
  • DETAILED DESCRIPTION OF THE INVENTION Figure 1 depicts an existing storage network system architecture.
  • the storage network 2 is connected to the server 1 and the disk storage system 4.
  • Each of the single disk storage systems (401, 402, 403) contains its own cache, and its cache can only be owned by a single disk storage system, but cannot be used by other disks. Storage system sharing.
  • Figure 1 depicts the system of the invention and its operating mechanism.
  • the storage network 2 is connected to a disk storage system 4 (three disk storage systems are shown in FIG. 2) and various servers 1.
  • the system of the present invention includes high-speed storage composed of a high-speed storage network cache module connected to the storage network 2.
  • the network cache system 5 and the storage network manager 3 composed of a storage network management module. Only two high-speed storage network cache modules are shown in this figure, which are 501 and 502 respectively.
  • the configuration can be configured in the form of data redundancy, configured as a mirror or Ra id5 (requires more than three high-speed storage network cache modules) to store data, so that data storage is guaranteed.
  • the figure also shows a storage network management module.
  • the high-speed storage network cache system 5 and the disk storage system 4 are connected through the storage network 2, and are composed of a hardware platform unit and a software platform unit, and the hardware platform unit adopts hardware resources of the embedded system or the server system.
  • the high-speed storage network cache system 5 is managed by the storage network manager 3, and can be shared and used by the server 1 and the storage system 4 connected to the storage network. Since the mechanical structure of the disk drive is not limited, and the cache can be shared, Therefore, the system of the present invention can respond to requests from servers at high speed.
  • FIG. 3 is a schematic structural diagram of a cache storage module of the high speed storage network of the present invention.
  • the high speed storage network cache module 8 is composed of a hardware platform unit 70 and a software platform unit 60, wherein the hardware platform unit 70 may be a server platform or an embedded system platform, having a central processing unit 705, interface hardware units 701 and 702, and a cache.
  • the hardware unit including the memory 703 or the PCI memory expansion unit 704) and the data guarantee unit (ie, the backup battery module 706).
  • the interface hardware unit can be an Ethernet port or a Fibre Channel port.
  • Each high-speed storage network cache module has at least two storage network interface hardware units, which are respectively used by interface software units in two modes.
  • the cache hardware unit can be implemented on the platform of the server platform or the embedded system by using the large-capacity memory of the two platforms, or can be implemented by the bus expansion cache module; the high-speed storage network cache module 8 is equipped as a data guarantee.
  • the backup battery module 706 of the unit can still maintain the data stored in the cache hardware unit for a certain period of time in the case that the external power supply is powered off, and enables the high-speed storage network cache module 8 to continue to operate, and the data is guaranteed to last. Can be transferred to the disk storage system.
  • the software platform unit 60 constituting the high speed storage network cache module 8 includes an interface software unit 601 operating in a target mode, an interface software unit 602 operating in a host mode, a cache management unit 603, and a data transfer software unit 604.
  • Two interface software units 601, 602 are provided for providing hardware interface drivers, such that the hardware interface units 701, 702 can provide Fibre Channel SCSI (FC SCSI) or Internet SCSI (iSCSI) based block level device interfaces; Level device interfaces (such as NFS and CIFS); can also provide object-based interfaces (such as 0SD).
  • FC SCSI Fibre Channel SCSI
  • iSCSI Internet SCSI
  • the interface software unit 601 in the target mode runs on the hardware interface unit 701, causing the high speed storage network cache module 8 to emulate a storage device that the server can read and write.
  • the interface software unit 602 in host mode runs on the hardware interface unit 702, so that the high speed storage network cache module acts as a server and can read and write other general purpose disk storage systems connected to the storage network.
  • the cache management unit 603 can use the high-end memory (non-system use memory) or map the memory extended by the PCI bus; the virtual memory file system (RAM Fi le System) or the virtual memory disk (RAM Disk) possessed by the operating system can be used.
  • the way to achieve, its function is: to map the data received from the interface software unit to the cache hardware unit (ie virtual file system or virtual memory disk).
  • the data transfer software unit 604 writes the data written from the server by the interface software unit in the host mode to the general disk storage system connected to the storage network through the interface hardware unit according to the final purpose of the write request.
  • the data transfer software unit is provided with a cache optimization module, which uses an optimization algorithm to integrate data sent to the disk storage system, and reduces the bandwidth occupied by the transferred data.
  • the specific method is: merge the repeated write request, and repeatedly write the request repeatedly. The last written data is written to the disk storage system, reducing the bandwidth used by the transferred data, and merging possible consecutive write requests, once to the disk storage system.
  • the processing flow of FIG. 3 is: The input and output request from the server is sent to the high speed storage network cache module 8 through the interface hardware unit 701, and processed by the target mode interface software unit 601; the interface software unit 601 sends the data to the cache management unit.
  • 603 The data is mapped into the memory by the cache management unit 603; the data transfer software unit 604 uses an optimization algorithm to write data to the other disk storage system through the host mode interface hardware unit 702 after the write data is mapped to the memory.
  • the high speed storage network cache system may be comprised of one or more high speed storage network cache modules 8.
  • a plurality of high-speed network connection to the storage network buffer storage module 8 may take the form of data redundancy configured in an array (such as mirroring or Raid5) 0
  • the high-speed storage network cache system can respond to the server's input and output requests at high speed. When processing random input and output requests, it has very high performance compared with the general disk storage system with ordinary hard disk as the storage medium, which can make the random request response rate. Improve hundreds of times.
  • the high speed storage network caching system is scalable. The management and scalability of the high-speed storage network cache system needs to be implemented in conjunction with the storage network manager.
  • the storage network manager includes a resource management unit, configured to manage resource information of the entire storage network, including addresses, capacities, and the like of the high-speed storage network cache system and the universal disk storage system; and a resource allocation unit that allocates a space of the high-speed network storage cache system; An expansion management unit for managing the expansion of the cache system of the high-speed storage network.
  • the storage network management module automatically tests changes in the storage network resources (newly added or reduced modules). And then the high speed storage network cache module is incorporated into the cache system according to a certain algorithm.
  • a data flow control unit configured to coordinate data flow control between the server and the high speed storage network cache system and the disk storage system, the data flow control unit allocates the flow direction of the data according to the request type, and the random high frequency request is cached through the high speed storage network.
  • the system accesses, and successive large requests are directly accessed through the disk storage system.
  • the storage network manager also includes an automatic test unit that stores the current status of all modules that may be connected to the storage network.
  • the timing (interval) sends commands to different devices, and the module determines whether the module is working properly or has been disconnected according to the response status. contact.
  • the server 1 acquires information of the storage network from the storage network manager 3, and the information is used to determine the flow direction of the data, and the random request is sent to the high-speed storage network cache system 5, and other requests are sent to the disk storage system. 4.
  • the high speed storage network cache system 5 and the disk storage system 4 are connected by the storage network 2, and the data written to the high speed storage network cache system 5 is finally transferred to the disk storage system 4.
  • both the high-speed storage network cache module 5 and the storage network manager 3 are mirrored.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Mathematical Physics (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Description

一种可扩充的高速存储网络緩存系统
技术领域 本发明是关于一种可扩充的高速存储网络緩存系统， 应用于存储网络 中， 由存储网络管理器管理。 背景技术 大容量和集中式的存储网络是存储技术发展的一个重要的趋势。 目前的 存储网络多用磁盘阵列设备作为通用存储系统，解决了存储容量和传输带宽 的问题。
由于硬磁盘机的机械结构的限制， 磁头寻道时间需要大约 10毫秒， 一 个硬磁盘机对随机请求的响应率仅仅约为每秒 100请求。 即使是网络存储使 用磁盘阵列的前提下， 把请求分配到各个磁盘上， 也很难突破每秒 1000请 求。
许多方案被提出来用于解决这个问题。 利用緩存技术是最常见的方法， 目前大部分緩存技术是基于服务器的， 或基于相应的磁盘存储系统， 如图 1 所示的緩存系统， 每个磁盘存储系统对应有自己的高速緩存。 但这些方案有 一定的局限性， 如緩存的容量有限制， 不能在多个磁盘存储系统之间共享緩 存等。 鉴于这些局限性， 有必要提出新的解决方案。 发明内容 为了克服现有技术的不足，本发明提供一种可扩充的高速存储网络緩存 系统，能突破现有对请求响应的极限，实现高速响应服务器的输入输出请求。
本发明的技术内容是： 一种可扩充的高速存储网络緩存系统， 其特征在 于包括连接在所述存储网络上的由至少一个高速存储网络緩存模块构成的 高速存储网絡緩存系统， 由至少一个存储网络管理模块构成的存储网络管理 器， 所述高速存储网络緩存模块由硬件平台单元和软件平台单元组成， 所述 硬件平台单元采用嵌入式系统平台或服务器平台， 所述高速存储网络緩存系 统由存储网络管理器管理，可供连接在存储网络上的服务器和存储系统共享 使用。 本发明由于在存储网络中接入了高速存储网络緩存系统， 而组成该系统 的高速存储网络緩存模块由硬件平台单元和软件平台单元组成，且硬件平台 单元是直接利用嵌入式系统平台或服务器平台硬件资源， 当配以相应的软件 平台单元后，使高速存储网络緩存系统成为在存储网络管理器管理下在存储 网络中能被所有服务器和存储系统共享使用的网络存储设备。这种网络设备 是虚拟成的高速緩存系统， 利用的是嵌入式系统平台或服务器平台的内存资 源或扩展内存资源， 而不是硬磁盘存储设备， 因此本发明大大突破了硬磁盘 机寻道緩慢的瓶颈问题， 实现了对服务器的输入输出请求的高速响应。
更进一步， 所述硬件平台单元包括中央处理器、 接口硬件单元、 高速緩 存硬件单元和数据保障单元， 高速緩存硬件单元在高速緩存软件单元的作用 下， 通过管理系统内部的内存或扩展内存， 把内存虚拟成为硬盘使用； 接口 硬件单元的设置，使该高速存储网络緩存系统可以被连入存储网络的服务器 和硬盘存储系统通过存储网络管理器的管理识别它并作为一个普通的高速 缓存存储器使用； 中央处理器的设置， 又使其具有读写其他存储器的服务功 能。 由于高速存储网络缓存系统与磁盘存储系统连通， 因此它又是一个 "透 明" 的存储设备， 最终数据将转移到磁盘存储系统上。 为使上述硬件平台单 元充分发挥作用， 设置了合理的软件平台单元， 如软件平台单元的数据转移 单元采用缓存优化模块，实现一定的优化算法，减少对存储网络带宽的占用。 再有， 网络存储管理器管理整个存储网络系统， 检测存储模块（包括高速存 储网络緩存模块和存储系统）扩充； 并且控制数据的流向， 随机数据的请求 发送给高速存储网络緩存系统 , 连续数据的请求直接发送给磁盘存储系统。 上述技术方案的设置能使随机请求的响应率可以大大改善。
因此， 高速存储网络緩存系统可以直接作为一个高性能的存储器给服务 器使用， 在网络存储管理器的控制下， 易于实现多个用户之间的空间共享、 空间动态分配以及緩存模块的动态容量增长。 附图说明 图 1是现有不能实现存储系统共享的緩存系统结构示意图。
图 2是本发明系统的结构示意图。
图 3是本发明高速存储网络緩存模块的结构框图。
图 4是本发明系统应用于存储网络的示意图。 具体实施方式 图 1描述了现有的存储网络系统架构。存储网络 2上连接有服务器 1和 磁盘存储系统 4 , 单个磁盘存储系统（401、 402、 403 )都包含有自己的高速 緩存， 其高速緩存只能被单个磁盘存储系统所有， 而不能被其他磁盘存储系 统共享。
图 1描述了本发明系统和其运行机理。存储网络 2上的连有磁盘存储系 统 4 (图 2中显示了三个磁盘存储系统）和各种服务器 1 , 本发明系统包括 连接在存储网络 2上的由高速存储网络緩存模块构成的高速存储网络緩存系 统 5和由存储网络管理模块构成的存储网絡管理器 3。 在此图中只显示了两 个高速存储网络緩存模块， 分别是 501、 502。 其配置可以釆用数据冗余的形 式构成阵列，配置成镜像或 Ra id5(需要 3个以上的高速存储网络緩存模块）， 以存放数据，使数据的存放得到保障。图中也显示了 1个存储网络管理模块。 高速存储网络緩存系统 5与磁盘存储系统 4通过存储网络 2连通，且由硬件 平台单元和软件平台单元组成， 所述硬件平台单元采用嵌入式系统或服务器 系统的硬件资源。 高速存储网络緩存系统 5由存储网络管理器 3管理， 可供 连接在存储网络上的服务器 1和存储系统 4共享使用， 由于不受磁盘机的机 械结构的限制， 而且能实现对緩存的共享, 因此本发明系统能高速响应服务 器的请求。 图 3显示了本发明高速存储网络緩存模块的结构示意图。 高速存储网络 緩存模块 8 , 由硬件平台单元 70和软件平台单元 60组成， 其中硬件平台单 元 70可以是基于服务器平台或嵌入式系统平台，具有中央处理器 705、接口 硬件单元 701和 702、 高速緩存硬件单元（包括内存 703或 PCI内存扩展单 元 704 )及数据保障单元（即后备电池模块 706 ) 中。 接口硬件单元可以是 以太网口 （Ethernet )或光纤通道口（f iber channe l) , 每个高速存储网络 緩存模块至少具备两个存储网络接口硬件单元，分别给两种模式下的接口软 件单元使用； 所述高速緩存硬件单元在服务器平台或嵌入式系统的平台上， 可以用两种平台的大容量内存实现， 也可以用总线扩展的緩存模块实现； 高 速存储网络緩存模块 8配备的作为数据保障单元的后备电池模块 706 , 在外 接电源断电的情况下，仍然能够在一定时间内保持储存在所述高速緩存硬件 单元内的数据， 并使得高速存储网络緩存模块 8能继续运行， 保障数据最后 能全部转移到磁盘存储系统。 构成高速存储网络緩存模块 8的软件平台单元 60包括工作在目标模式 下的接口软件单元 601、 工作在主机模式下的接口软件单元 602、 高速緩存 管理单元 603、 数据转移软件单元 604。 两个接口软件单元 601、 602用于提 供硬件接口驱动， 使硬件接口单元 701、 702可提供基于光纤通道 SCSI ( FC SCSI )或网际网络 SCSI ( iSCSI ) 的数据块级设备接口； 也可以提供文件级 设备接口 （如 NFS和 CIFS ); 也可以提供基于对象的接口 （如 0SD )。 目标模 式下的接口软件单元 601在硬件接口单元 701上运行，使高速存储网络緩存 模块 8模拟一个存储设备， 服务器可以读写这个存储设备。 主机模式下的接 口软件单元 602在硬件接口单元 702上运行，使高速存储网络緩存模块如同 一个服务器， 可以读写其他连接在存储网络上的通用磁盘存储系统。 高速緩 存管理单元 603可以利用高端内存（非系统使用内存）或映射 PCI总线扩展 的内存的方式实现； 可采用操作系统具有的虚拟内存文件系统 （RAM Fi le System )或虛拟内存盘（RAM Disk )的方式实现， 它的功能是： 把从接口软 件单元所收到的数据， 映射到所述高速緩存硬件单元（即虛拟文件系统或虚 拟内存盘上）。 所述数据转移软件单元 604根据写入请求的最终目的， 把从 服务器写入的数据， 由主机模式下的接口软件单元通过接口硬件单元， 写入 到连接在存储网络上的通用磁盘存储系统。 所述数据转移软件单元设置有 緩存优化模块， 该模块采用优化算法， 整合发送到磁盘存储系统的数据， 减 少转移数据占用的带宽， 具体的做法是： 合并重复写请求， 把多次重复写请 求最后一次写的数据写入磁盘存储系统， 减少转移数据占用的带宽， 且合并 可能的连续写请求， 一次写入到磁盘存储系统中。
图 3 的处理流程为： 来自于服务器的输入输出请求通过接口硬件单元 701送到高速存储网络緩存模块 8 , 由目标模式接口软件单元 601进行处理； 接口软件单元 601把数据发送给高速緩存管理单元 603; 由高速緩存管理单 元 603把数据映射到内存中； 数据转移软件单元 604采用优化算法， 在写入 数据映射到内存后通过主机模式接口硬件单元 702把数据写入其它的磁盘存 储系统中。 高速存储网络緩存系统可由一个或多个高速存储网络緩存模块 8构成。 同时连接到存储网络的多个高速存储网络緩存模块 8可以采用数据冗余的形 式构造成阵列 （如镜像或 Raid5 )0 高速存储网络緩存系统可以高速响应服务器的输入输出请求，在处理随 机输入输出请求时， 与以普通硬磁盘为存储介质的通用磁盘存储系统相比 , 具有非常高的性能， 能使随机请求的响应率改善上百上千倍。 高速存储网络 緩存系统具有可扩充性。 高速存储网絡缓存系统的管理和扩充性， 需要结合 存储网络管理器来实现。 存储网絡管理器中包括资源管理单元， 用于管理整 个存储网络的资源信息， 包括高速存储网络緩存系统和通用磁盘存储系统的 地址、 容量等； 资源分配单元， 分配高速网络存储緩存系统的空间； 扩充管 理单元， 用于管理高速存储网络缓存系统的扩充， 当存储网络增加或减少一 个高速存储网络緩存模块时，存储网络管理模块将自动测试存储网络资源的 变化（新增加的或减少的模块）， 然后根据一定的算法将该高速存储网络緩 存模块并入緩存系统。 数据流控制单元， 用于协调服务器到高速存储网络緩 存系统和磁盘存储系统之间的数据流控制，数据流控制单元根据请求类型来 分配数据的流向， 随机的高频率的请求通过高速存储网络緩存系统来存取， 连续的大的请求通过磁盘存储系统直接存取。
存储网络管理器还包括一个自动测试单元，存有当前所有可能连接在存 储网络的模块的状态， 定时（间隔一段时间）给不同的设备发送命令， 根据 响应状态判断该模块工作正常或已经断开联系。 图 4所示的存储网络示意图中服务器 1从存储网絡管理器 3获取存储网 络的信息， 由这些信息来决定数据的流向， 随机请求发送到高速存储网络緩 存系统 5 , 其它请求发送到磁盘存储系统 4。 高速存储网络緩存系统 5和磁 盘存储系统 4之间通过存储网络 2来连接， 写入高速存储网络緩存系统 5的 数据最终将转移到磁盘存储系统 4。 本图中高速存储网络緩存模块 5和存储 网络管理器 3均采取镜像设置。

Claims

权利要求
1、 一种可扩充的高速存储网络緩存系统， 其特征在于包括连接在存储 网络上的由至少一个高速存储网络緩存模块构成的高速存储网络緩存系統 和连接在存储网络上的由至少一个存储网络管理模块构成的存储网络管理 器， 所述高速存储网絡緩存模块由硬件平台单元和软件平台单元组成， 所述 硬件平台单元采用嵌入式系统平台或服务器平台， 所述高速存储网络緩存系 统由存储网络管理器管理，可供连接在存储网络上的服务器和磁盘存储系统 共享使用。
2、 根据权利要求 1所述的可扩充的高速存储网络緩存系统， 其特征在 于所述硬件平台单元包括中央处理器、 接口硬件单元、 高速缓存硬件单元和 数据保障单元。
3、 根据权利要求 2所述的可扩充的高速存储网络緩存系统， 其特征在 于所述接口硬件单元为各种存储网络硬件接口， 包括光纤通道接口和以太网 接口
4、 根据权利要求 2所述的可扩充的高速存储网络緩存系统 , 其特征在 于所述高速緩存硬件单元采用服务器平台或嵌入式系统平台的大容量内存， 或采用两种平台的总线扩展的内存。
5、 根据权利要求 2所述的可扩充的高速存储网絡緩存系统， 其特征在 于所述数据保障单元为配置在所述硬件平台单元内的后备电池模块。
6、 根据权利要求 1所述的可扩充的高速存储网络緩存系统， 其特征在 于所述软件平台单元包括工作在目标模式下的接口软件单元、工作在主机模 式下的接口软件单元、 高速緩存管理单元、 数据转移软件单元， 所述两个接 口软件单元用于提供硬件接口驱动 , 所述高速緩存管理单元用于把从接口软 件单元所收到的数据， 映射到所述高速緩存硬件单元， 所述数据转移软件单 元根据写入请求的最终目的， 把从服务器写入的数据， 由主机模式下的接口 软件单元通过接口硬件单元， 写入到连接在存储网络上的存储系统， 所述 数据转移软件单元设置有緩存优化模块， 该模块采用优化算法， 整合发送到 存储系统的数据， 减少转移数据占用的带宽。
7、 根据权利要求 1 所述的可扩充的高速存储网络緩存系统， 其特征在 于所述存储网络管理模块包括资源管理单元， 用于管理整个存储网络的资源 信息； 资源分配单元， 用于分配高速存储网络緩存系统的空间给各个服务器 和 /或磁盘存储系统； 扩充管理单元， 用于管理高速存储网络緩存系统的扩 充； 数据流控制单元， 用于协调服务器到高速存储网络緩存系统和磁盘存储 系统之间的数据流控制。
8、 根据权利要求 7所述的可扩充的高速存储网络緩存系统， 其特征在 于数据流控制单元根据请求类型来分配数据的流向。 随机数据的请求写入高 速存储网络缓存系统， 连续数据的请求写入磁盘存储模块。
9、 根据权利要求 1 所述的可扩充的高速存储网络緩存系统， 其特征在 于所述存储网络管理模块为两台， 采用镜像配置， 所述高速存储网络緩存模 块为两台或两台以上， 配置成冗余阵列的形式。
PCT/CN2006/000384 2005-03-14 2006-03-14 Systeme tampon de stockage en reseau expansible haut debit WO2006097037A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US11/886,435 US8032610B2 (en) 2005-03-14 2006-03-14 Scalable high-speed cache system in a storage network

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200510011424.9 2005-03-14
CNB2005100114249A CN100342352C (zh) 2005-03-14 2005-03-14 一种可扩充的高速存储网络缓存系统

Publications (1)

Publication Number Publication Date
WO2006097037A1 true WO2006097037A1 (fr) 2006-09-21

Family

ID=35035896

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2006/000384 WO2006097037A1 (fr) 2005-03-14 2006-03-14 Systeme tampon de stockage en reseau expansible haut debit

Country Status (3)

Country Link
US (1) US8032610B2 (zh)
CN (1) CN100342352C (zh)
WO (1) WO2006097037A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102158344A (zh) * 2011-05-20 2011-08-17 苏州安源汇信软件有限公司 并行组播网络文件系统
US8032610B2 (en) 2005-03-14 2011-10-04 Yaolong Zhu Scalable high-speed cache system in a storage network

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7707248B2 (en) * 2007-06-25 2010-04-27 Microsoft Corporation Credit-based peer-to-peer storage
CN101540685B (zh) * 2008-06-06 2012-08-29 曙光信息产业(北京)有限公司 一种用于刀片服务器的PCIe共享存储刀片
JP5187017B2 (ja) * 2008-06-18 2013-04-24 富士通株式会社 分散ディスクキャッシュシステム及び分散ディスクキャッシュ方法
US20100088520A1 (en) * 2008-10-02 2010-04-08 Microsoft Corporation Protocol for determining availability of peers in a peer-to-peer storage system
EP2403202A1 (en) 2010-07-02 2012-01-04 Thomson Licensing Computer system and method for operating the same
CN103036958B (zh) * 2012-12-06 2016-02-24 陕西维德科技股份有限公司 一种服务器集群高速存储系统
CN104426838B (zh) * 2013-08-20 2017-11-21 中国移动通信集团北京有限公司 一种互联网缓存调度方法及系统
CN104580479A (zh) * 2015-01-15 2015-04-29 浪潮集团有限公司 一种san存储系统主机链路带宽认证方法
CN105677248A (zh) * 2015-12-31 2016-06-15 华为技术有限公司 一种数据访问方法和装置
CN106227473A (zh) * 2016-09-27 2016-12-14 天脉聚源(北京)传媒科技有限公司 一种将内存作为磁盘读写的实现方法和装置
US10659555B2 (en) * 2018-07-17 2020-05-19 Xilinx, Inc. Network interface device and host processing device
US10838763B2 (en) 2018-07-17 2020-11-17 Xilinx, Inc. Network interface device and host processing device
CN111212141A (zh) * 2020-01-02 2020-05-29 中国科学院计算技术研究所 一种共享存储系统

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1138216C (zh) * 2000-06-21 2004-02-11 国际商业机器公司 为多种设备提供快速信息服务的装置及方法
US6862606B1 (en) * 2001-05-11 2005-03-01 Novell, Inc. System and method for partitioning address space in a proxy cache server cluster

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19509262C2 (de) 1995-03-15 2001-11-29 Siemens Ag Halbleiterbauelement mit Kunststoffumhüllung und Verfahren zu dessen Herstellung
US6640278B1 (en) * 1999-03-25 2003-10-28 Dell Products L.P. Method for configuration and management of storage resources in a storage network
US6523102B1 (en) * 2000-04-14 2003-02-18 Interactive Silicon, Inc. Parallel compression/decompression system and method for implementation of in-memory compressed cache improving storage density and access speed for industry standard memory subsystems and in-line memory modules
US20010049773A1 (en) * 2000-06-06 2001-12-06 Bhavsar Shyamkant R. Fabric cache
US6792507B2 (en) * 2000-12-14 2004-09-14 Maxxan Systems, Inc. Caching system and method for a network storage system
US7472231B1 (en) * 2001-09-07 2008-12-30 Netapp, Inc. Storage area network data cache
EP1298878A1 (en) * 2001-09-26 2003-04-02 Telefonaktiebolaget L M Ericsson (Publ) Hierarchical caching in telecommunication networks
JP2003131946A (ja) * 2001-10-19 2003-05-09 Nec Corp キャッシュメモリ制御装置及び方法
US7007142B2 (en) * 2002-02-19 2006-02-28 Intel Corporation Network data storage-related operations
WO2003075166A1 (fr) * 2002-03-06 2003-09-12 Fujitsu Limited Systeme de stockage et procede de transfert de donnees dans ledit systeme
US7043610B2 (en) * 2002-08-19 2006-05-09 Aristos Logic Corporation System and method for maintaining cache coherency without external controller intervention
JP4439798B2 (ja) * 2002-10-17 2010-03-24 株式会社日立製作所 ディスクアレイ装置の制御方法、及びディスクアレイ装置
JP4278445B2 (ja) * 2003-06-18 2009-06-17 株式会社日立製作所 ネットワークシステム及びスイッチ
US7437511B1 (en) * 2003-06-30 2008-10-14 Storage Technology Corporation Secondary level cache for storage area networks
CN1310483C (zh) * 2003-08-19 2007-04-11 广东省电信有限公司科学技术研究院 保证端到端ip电信服务质量的网络系统和控制方法
US8549226B2 (en) * 2004-05-14 2013-10-01 Hewlett-Packard Development Company, L.P. Providing an alternative caching scheme at the storage area network level
CN100342352C (zh) 2005-03-14 2007-10-10 北京邦诺存储科技有限公司 一种可扩充的高速存储网络缓存系统
US7415574B2 (en) * 2006-07-05 2008-08-19 Cisco Technology, Inc. Dynamic, on-demand storage area network (SAN) cache

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1138216C (zh) * 2000-06-21 2004-02-11 国际商业机器公司 为多种设备提供快速信息服务的装置及方法
US6862606B1 (en) * 2001-05-11 2005-03-01 Novell, Inc. System and method for partitioning address space in a proxy cache server cluster

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8032610B2 (en) 2005-03-14 2011-10-04 Yaolong Zhu Scalable high-speed cache system in a storage network
CN102158344A (zh) * 2011-05-20 2011-08-17 苏州安源汇信软件有限公司 并行组播网络文件系统

Also Published As

Publication number Publication date
CN100342352C (zh) 2007-10-10
CN1664794A (zh) 2005-09-07
US8032610B2 (en) 2011-10-04
US20080172489A1 (en) 2008-07-17

Similar Documents

Publication Publication Date Title
WO2006097037A1 (fr) Systeme tampon de stockage en reseau expansible haut debit
US6754785B2 (en) Switched multi-channel network interfaces and real-time streaming backup
CN101393536B (zh) 存储系统
US7272687B2 (en) Cache redundancy for LSI raid controllers
US7149846B2 (en) RAID protected external secondary memory
US7669019B2 (en) Apparatus and method of mirroring data between nonvolatile memory and hard disk
US6981070B1 (en) Network storage device having solid-state non-volatile memory
JP2007079958A (ja) 記憶制御装置、データ処理方法、及びコンピュータプログラム
CN101566927B (zh) 存储系统和存储控制器以及数据缓存方法
CN1723434A (zh) 用于可扩展网络附接存储系统的装置和方法
KR20020012539A (ko) 공유된 디스크 어레이 관리 기능을 구현하기 위한 방법 및시스템
US9785561B2 (en) Integrating a flash cache into large storage systems
EP2422272B1 (en) Active-active support of virtual storage management in a storage area network ("san")
US7162582B2 (en) Caching in a virtualization system
US7725654B2 (en) Affecting a caching algorithm used by a cache of storage system
US20030200389A1 (en) System and method of cache management for storage controllers
KR101564712B1 (ko) Ｓｃｓｔ를 이용한 총 플래시 어레이 스토리지 가상화 방법
JP4640854B2 (ja) ディスクアレイ制御方法及びディスクアレイ制御装置
He et al. STICS: SCSI-to-IP cache for storage area networks
JP3668922B2 (ja) システム
JP4111450B2 (ja) ディスクアレイ制御方法及びディスクアレイ制御装置
US20230009942A1 (en) Using drive compression in uncompressed tier
CN118502666A (zh) 支持分层分级大容量扩展的高速数据存储架构与读写方法
Wang et al. Disk Tree: A Fat-Tree Based Heterogeneous Multi-tier Storage Architecture
Chen et al. Design of redundant array of independent DVD libraries based on iSCSI

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application
NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 11886435

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: RU

WWW Wipo information: withdrawn in national office

Country of ref document: RU

122 Ep: pct application non-entry in european phase

Ref document number: 06722040

Country of ref document: EP

Kind code of ref document: A1

WWW Wipo information: withdrawn in national office

Ref document number: 6722040

Country of ref document: EP