Architecture for providing block-level storage access over a computer network

Abstract

A network-based storage system comprises one or more block-level storage servers that connect to, and provide disk storage for, one or more host computers (“hosts”) over logical network connections (preferably TCP/IP sockets). In one embodiment, each host can maintain one or more socket connections to each storage server, over which multiple I/O operations may be performed concurrently in a non-blocking manner. The physical storage of a storage server may optionally be divided into multiple partitions, each of which may be independently assigned to a particular host or to a group of hosts. Host driver software presents these partitions to user-level processes as one or more local disk drives. When a host initially connects to a storage server in one embodiment, the storage server initially authenticates the host, and then notifies the host of the ports that may be used to establish data connections and of the partitions assigned to that host.

Description

APPENDICES

[0002] This specification includes appendices A-D which contain details of a commercial implementation of the invention. The appendices are provided for illustrative purposes, and not to define or limit the scope of the invention.

BACKGROUND OF THE INVENTION

[0003] 1. Field of the Invention

[0004] The present invention relates to storage systems for computer networks, and more specifically, relates to software architectures for providing block level access to storage resources on a network.

[0005] 2. Description of the Related Art

[0006] Various types of architectures exist for allowing host computers to share hard disk drives and other storage resources on a computer network. One common type of architecture involves the use of a central file manager. One problem with this architecture is that the failure of the central file manager can render the entire system inoperable. Another problem is that many software applications are not designed to use a central file manager.

[0007] Some storage architectures overcome these deficiencies by allowing the host computers to access the storage resources directly over the network, without the use of a central file manager. Typically, these architectures allow the host to access the storage resources over a network connection at the block level (as opposed to the file level). One problem with this type of architecture is that the failure of an input/output request can cause other pending requests from the same host to be delayed. Another problem is that the architecture is highly vulnerable to network failures. The present invention addresses these and other problems.

SUMMARY OF THE INVENTION

[0008] The present invention comprises a system architecture for providing block-level access to storage resources, such as disk arrays, over a computer network without the need for a central file manager. The architecture embodies various inventive features that may be implemented individually or in combination.

[0009] One feature of the architecture is that concurrent input/output (I/O) requests from the same host computer (“host”) are handled over separate logical network connections or sockets (preferably TCP/IP sockets). For example, in a preferred embodiment, a given host can establish two socket connections with a given block-level storage server, and use one socket to perform one I/O request while using the other socket to perform another I/O request. As a result, the failure or postponement of one I/O request does not block or interfere with other I/O requests.

[0010] Another feature of the architecture is that the sockets can be established over multiple networks, including networks of different types and bandwidths, to provide increased fault tolerance. For example, a given host computer and storage server can be connected by two networks that support the TCP/IP protocol, one of which may provide a much lower transfer rate than the other. As long as one of these networks is functioning properly, the host will be able to establish a logical connection to the block server and execute I/O requests.

[0011] In one embodiment, the architecture includes a host-side device driver and a host-side reader/writer component that run on the host computers. The architecture also includes a server-side device driver and a server-side reader/writer component that run on the block-level storage servers. The reader/writer components are preferably executed as separate processes that are established in pairs (one host-side reader/writer process and one server-side reader/writer process), with each pair dedicated to a respective socket over a network. For example, if two logical connections are established between a given host computer and a given storage server, each such socket will be managed by a different pair of reader/writer processes. The reader/writer processes and sockets preferably remain persistent over multiple I/O requests. The device drivers and reader/writer processes operate to export the block-level-access interface of the storage servers to the host computers, so that the disk drives of the block servers appear to the host computers as local storage resources.

[0012] In accordance with one inventive feature of the architecture, when an I/O request from a host process involves the transfer of more than a threshold quantity of data, the host's device driver divides the I/O requests into two or more constituent I/O operations. Each such operation is assigned to a different socket connection with the target storage server such that the constituent operations may be performed, and the associated I/O data transferred, in parallel over the network. This feature of the architecture permits relatively large amounts of network bandwidth to be allocated to relatively large I/O requests.

[0013] Another feature of the architecture is a mechanism for dividing the physical storage space or units of a block-level storage server into multiple partitions, and for allocating these partitions to hosts independently of one another. In a preferred embodiment, a partition can be allocated uniquely to a particular host, or can be allocated to a selected group of hosts (in which case different hosts may have different access privileges to the partition). The partition or partitions assigned to a particular host appear, and can be managed as, one or more local disk drives.

[0014] Yet another inventive feature of the architecture is an authentication and discovery protocol through which a storage server authenticates a host, and then provides access information to the host, before permitting the host to access storage resources. In a preferred embodiment, when the host is booted up, it initially establishes a configuration socket connection to the storage server. Using this configuration socket, the storage server authenticates the host—preferably using a challenge-response method that is dependent upon a version of the driver software. If the authentication is successful, the storage server provides access information to the host, such as the identities of dynamic ports which may be used by the host for data connections to the storage server, and information about any partitions of the storage server that are assigned to that host. This feature of the architecture provides a high degree of security against unauthorized accesses, and allows storage partitions to be securely assigned to individual hosts.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] These and other features will now be described with reference to the drawings of certain embodiments of the invention, which are intended to illustrate, and not limit, the scope of the invention.

[0016] FIG. 1 illustrates the primary hardware components of an example system in which the invention may be embodied, including a host computer and a block server.

[0017] FIG. 2 illustrates the software architecture of the system of FIG. 1, including host-side and server-side device drivers and reader/writer (RW) components that operate according to the invention.

[0018] FIG. 3 illustrates examples of the types of networks and network components that can be used to interconnect the hosts and block servers.

[0019] FIG. 4 shows, in example form, how the concurrent socket connections are established between pairs of reader/writer components.

[0020] FIG. 5 illustrates the flow of information between components when a host computer performs a read from a block server.

[0021] FIG. 6 illustrates the flow of information between components when a host computer performs a write to a block server.

[0022] FIG. 7 illustrates how I/O requests are assigned to socket connections transparently to user-level applications, and illustrates how an I/O request may be subdivided for processing over multiple TCP/IP connections.

[0023] FIG. 8 illustrates how the physical storage of a block server may be divided into multiple partitions, each of which may be independently allocated to one or more host computer.

[0024] FIG. 9 illustrates an authentication and discovery protocol through which a host computer is authenticated by a block server, and then obtains information for accessing the block server.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0025] The system architecture described in this section, and in the attached appendices, embodies various inventive features that may be used individually or in combination. Some of these features may be implemented without others, and/or may be implemented differently than set forth herein, without departing from the scope of the invention as defined by the appended claims.

[0026] I. Overview

[0027] The present invention comprises a system architecture for providing block-level storage access over one or more computer networks. The architecture is designed to incorporate any number of host computers and block-level storage servers communicating across a network or a combination of networks. In one embodiment, the architecture exports virtualized storage blocks over TCP/IP connections. Because TCP/IP is used for communications between the host computers and block-level storage servers in a preferred embodiment, a variety of network topologies can be used to interconnect the host computers and the block servers of a given system. For example, for relatively small systems, the host computers and storage servers can be interconnected by a hub, while for larger systems, the hub may be replaced with a switch.

[0028] Depicted in FIG. 1 are the hardware components of a typical system that embodies the invention. The system includes a host computer 102 (“host”) and a block-level IP storage server 104 (“block server”) interconnected by a network 100 via respective network interface cards 106, such as 10/100/1000 Base-T or 1000 Base-SX Gigabit Ethernet cards. The host computer 102 may be a standard PC or workstation configured to operate as a server or as a user computer. The block server 104 may be a network-attached IP storage box or device which provides block-level data storage services for host computers 102 on the network 100.

[0029] In the illustrated embodiment, the block server 104 includes a disk array controller 110 that controls an array of disk drives 112. A disk array controller 110 of the type described in U.S. Pat. No. 6,098,114 may be used for this purpose, in which case the disk drives 112 may be ATA/IDE drives. The disk array controller may support a variety of disk array configurations, such as RAID 0, RAID 5, RAID 10, and JBOD, and is preferably capable of processing multiple I/O requests in parallel. The block server 104 also includes a CPU board and processor 108 for executing device drivers and related software. The block server may also include volatile RAM (not shown) for caching I/O data, and may include flash or other non-volatile solid state memory for storing configuration information (see FIG. 8).

[0030] In one embodiment, the network 100 may be any type or combination of networks that support TCP/IP sockets, including but not limited to Local Area Networks (LANs), wireless LANs (e.g., 802.11 WLANs), Wide Area Networks (WANs), the Internet, and direct connections. One common configuration is to locally interconnect the hosts 102 and block servers 104 by an Ethernet network to create an Ethernet-based SAN (Storage Area Network). As depicted by dashed lines in FIG. 1, the host and the block server 102, 104 may be interconnected by a second network 100′, using a second set of network cards 106′, to provide increased fault tolerance (as described below). The two networks 100, 100′ may be disparate networks that use different mediums and provide different transfer speeds. Some of the various network options are described in more detail below with reference to FIG. 3.

[0031] The software components of the architecture are shown in FIG. 2. The host side 102 of the software architecture includes an operating system (O/S) 202 such as Unix, Windows NT, or Linux; a host-side device driver 204 (“host driver”) which communicates with the operating system 202; and a reader/writer (RW) component 200a (also referred to as an “agent”) which communicates with the host driver 204. The storage side 104 of the software architecture includes a reader/writer (RW) component 200b and a storage-side device driver 206 (“server driver”) that are executed by the CPU board's processor 108 (FIG. 1). The server driver 206 initiates disk operations in response to I/O requests received from the server-side RW component 200b.

[0032] The RW components 200a, 200b are preferably executed as separate processes that are established in pairs (one host-side RW process and one server-side RW process), with each pair dedicated to a respective TCP/IP socket over a network 100. The host RW 200a operates generally by “reading” I/O requests from the host driver 204, and “writing” these requests onto the network 100. Similarly, the storage RW 200b operates generally by reading I/O requests from the network 100 and writing these requests to the server driver 206. This process can occur simultaneously with transfers by other RW pairs, and can occur in any direction across the network 100. The RW components 200 also preferably perform error checking of transferred I/O data.

[0033] Each RW process (and its corresponding socket) preferably remains persistent on its respective machine 102, 104, and processes I/O requests one at-a-time on a first-in-first-out basis until the connection fails or is terminated. A host computer 102 establishes a socket by sending a service request over a dedicated configuration socket to the relevant block server 104. Once a socket connection is established between a RW pair 200a, 200b, the socket handles bi-directional traffic between the host computer 102 and block server 104.

[0034] In the illustrated embodiment, the RW components 200 run as processes that are separate from the host and server drivers 204, 206, respectively. The host-side 200a and storage-side 200b RW could alternatively be implemented, for example, as one or more of the following: (a) part of the host and server drivers 204, 206 (respectively), (b) separate device drivers 204, 206 (respectively), (c) separate kernel threads, (d) multiple threads within a single process, (e) multiple threads within multiple processes, and (f) multiple processes within a single thread.

[0035] A host computer 102 may establish multiple logical connections (sockets) to a given block server 104, and/or establish sockets to multiple different block servers 104 (as discussed below). An important benefit of this feature is that it allows multiple I/O requests from the same host to be processed concurrently (each over a separate socket) in a non-blocking manner—if one socket fails, the I/O requests being performed over other sockets are not affected. Each socket is managed by a respective RW pair.

[0036] An important function of the host driver 204 is that of virtualizing the storage provided by the block servers 204, so that all higher-level software processes on the host, such as the operating system and other user-level processes, view the block server storage as one or more local, physical disk drives. To accomplish this task, the host driver dynamically assigns I/O requests to TCP/IP socket connections without revealing the existence of such connections, or any other network details, to user-level processes. The block server 104 preferably appears to the host's user-level processes as a SCSI device, allowing conventional volume managers to be used.

[0037] As described below in sub-section III, one embodiment of the architecture permits the physical storage of a block server 104 to be divided into multiple, variable-size partitions. Each such partition may be independently allocated to one or more hosts, and may configured such that it is viewed and managed as a separate physical disk drive. In other embodiments, block-level access may be provided to the hosts without partitioning.

[0038] FIG. 3 shows some of the various networks 100 and network components that may be used to interconnect the host 102 and block servers 104 of a given system. These include a hub 302 (commonly used to connect LAN segments), the Internet 304, a router 306 (a computer that forwards packets according to header information), a switch 308 (a device that filters and forwards packets between LAN segments), and a gateway 310 (a computer that interconnects two different types of networks). The system architecture allows any combination of these network options to be used to interconnect a given host computer 102 and block server 104.

[0039] An important feature of the architecture is that when the network 100 becomes inundated with traffic, a network 100 administrator can either add network 100 capabilities on the fly or change the network 100 hardware without causing any loss of data. The host-side 102 and storage-side 104 software components are configured, using conventional methods, to detect and use new network 100 connections as they become available, and to retry operations until a connection is established. For example, a network 100 administrator could initially connect thirty host computers 102 to a small number of block servers 104 using a network hub 302. When the number of computers reaches a level at which the network hub 302 is no longer suitable, a 1000-port switch could be added to the network 100 and the hub 302 removed without taking the network 100 off-line. The architecture functions this way because the host RW 200a creates a new sockets connection to the storage RW 200b automatically as new physical connections become available.

[0040] The architecture and associated storage control protocol present the storage resources to the host computers 102 as a logically contiguous array of bytes which are accessible in blocks (e.g., of 512 bytes). The logical data structures of the implementation support byte level access, but disk drives typically export blocks which are of a predetermined size, in bytes. Thus, to access a given block, a block address (sector number) and a count of the number of blocks (sectors) is provided. In one embodiment, the protocol exports a 64-bit logical block address (LBA) and 64-bit sector count. On write operations, the I/O write data request is packaged into a block structure on the host side 102. The block request and data are sent to the block server 104 over one or more of the socket connections managed by the host RW processes 200a. The architecture also allows data to be stored non-sequentially and allows for the storage medium to efficiently partition space and reclaim unused segments.

[0041] Depicted in FIG. 4 are sample socket connections 400 made by RW pairs 200 connecting over a network 100 to link host computers 102 to block servers 104. As mentioned above, the network 100 may actually consist of multiple networks 100, including fully redundant networks 100. Each host computer 102 can open one or more socket connections 400 (using corresponding RW pairs) to any one or more block servers 104 as needed to process I/O requests. New socket connections 400 can be opened, for example, in response to long network 100 response times, failed socket connections 400, the availability of new physical connections, and increases in I/O requests. For example, a host computer 102 can initially open two sockets 400 to a first block server 104; and subsequently open two more sockets 400 to another block server 104 as additional storage resources are needed. Another host computer 102 may have open socket connections 400 to the same set of block servers 104 as shown. As described above, each socket 400 acts as an independent pipeline for handling I/O requests, and remains open until either an error occurs or the host 102 terminates the socket connection 400.

[0042] II. Processing of Input/Output Requests

[0043] FIGS. 5 and 6 illustrate a network storage protocol that may be used for I/O read operations and write operations (respectively) between a host computer 102 and a block server 104 over a socket connection 400. Located at the tops of the vertical lines in FIGS. 5 and 6 are abbreviations that denote components as follows.

[0044] OS Operating System

[0045] HD=Host Driver 204

[0046] HRW=Host Computer's Reader/Writer 200a

[0047] N=Network

[0048] SRW=Server Reader/Writer 200b (of block server)

[0049] SD=Server Driver 206 (of block server)

[0050] Time increases, but is not shown to scale, in these diagrams moving from top to bottom. Arrows from one vertical line to another generally represent the flow of messages or data between components. An arrow that begins and ends at the same component (vertical line) represents an action performed by that component. The small circles in the figures represent rendezvous events.

[0051] In one embodiment, as shown in FIG. 5, the host reader/writer (HRW) initially sends a request 1H to the host driver (HD) for an I/O command packet, indicating that the socket is available for use. This step can be viewed as the message “if you have work to do, give it to me.” The host driver eventually responds to this request by returning a command packet that specifies an I/O request, as shown. As represented by the arrow labeled 2H, the host reader/writer (HRW) translates the command packet into a network-generalized order. This step allows different, cross platform, computer languages to function on a common network 100. The local computational transformation of a host command packet, or host language, to a network command packet, or network language, is architecture specific.

[0052] At this point, the host reader/writer (HRW) generates two networks events, 3H and 4H. Message 4H represents a post of a received network response packet, from 3H, across the network 100 and is discussed below. Message 3H represents the network generalized command packet being written over a pre-existing “pinned-up” TCP/IP connection. In order for this transfer to occur in the preferred embodiment, a rendezvous must take place with 1S, which represents a network 100 request to receive the command packet. This request 1S has the ability to wait indefinitely if there is no “work” to be done. Once the network 100 rendezvous is satisfied, the command packet is received by the block server's reader/writer (SRW), and is re-translated by the SRW to the server-side language via step 2S. Step 2S is similar to the translation of the host command packet to a network command packet shown in 2H.

[0053] As further illustrated in FIG. 5, message 3S represents the server-side reader/writer posting the command packet to the server driver (SD) 206. Included in the command packet are the following: an I/O unit number (a small integer that is a logical identifier for the underlying disk drive partition on any form of storage disks), a command (a small integer indicating the type of command, such as a read operation or a write operation), a starting logical block address (an integer indicating the starting block or sector for the I/O operation), and the block count (an integer indicating the number of blocks or sectors for the I/O operation).

[0054] After the command packet is delivered to the server device driver (SD), a response is sent back to the server-side reader/writer (SRW). As depicted by 4S, the server-side reader/writer transforms this response packet from storage-side order to network order via step 4S. Subsequently, as depicted by 5S and 6S, both a network write of the response packet (5S) and a network write of the I/O data from the read (6S) occur simultaneously.

[0055] A rendezvous thereafter occurs between the response packet (5S) and the network read request (4H), representing satisfaction of the network read request. This step is performed for two reasons: (1) to determine whether the network response packet contains completion status after it has been checked for errors, and (2) if the read operation failed, to prevent the requested data from being returned to the host 102. After they meet, the network response packet is translated into the host response packet language in 5H.

[0056] Next, there is a network read request 6H for the I/O (read) data. This request 6H is satisfied by the network write 6S of the I/O read data, causing the I/O read data to be transferred to the host computer. Finally, in 7H, the host reader/writer notifies the host driver, and the host driver notifies the operating system, that the I/O operation is complete.

[0057] With regard to the write operations shown in FIG. 6, the steps are similar up through 3S. The description below will show the remaining steps with regard to the remaining I/O data transfers and network 100 operations.

[0058] The write data associated with this I/O operation is received by the network 100 and a receive operation is posted by the block server 104 to the network 100 in step 3S. This network read occurs subsequent to the translation of the network command packet to storage system language in step 2S. This ensures that a network read operation is posted for the proper size subsequent to error checking of the network command packet. In step 4S, the block server 104 sends the storage system command packet to the server driver (SD). This command packet contains the same type of information as described for the read case. The SD returns a response packet subsequent to the logical (or physical) completion of the specified write operation which is subsequently converted to network order in step 5S. This network response packet is sent to the host computer 102 via a send to network command in step 6S. This network response packet is received by the host computer 102 and translated to a host response packet in step 6H. Subsequently, the host computer 102 sends a command to the HD to complete the I/O operation. Steps 6H and 7H are similar to the steps of the read operation in FIG. 5 in which the host reader/writer notifies the host driver, and the host driver notifies the operating system, that the I/O operation is complete.

[0059] Appendix A and B include additional details of the host and server side logic according to one embodiment of the invention. In addition, Appendix C specifies implementation details of one embodiment of the protocol by which the hosts 102 and block servers 104 communicate. Certain portions of these appendices correspond to prototype implementations in which only some of the system's functions are enabled.

[0060] FIG. 7 illustrates how the I/O requests from user-level applications are dynamically assigned to TCP/IP socket connections transparently to these applications, and thus without the need for a fixed mapping between applications and sockets. The drawing also illustrates how, in one embodiment, an I/O request can be subdivided into smaller operations which can be issued and performed in parallel over multiple socket connections. In this illustration, the host-side RW components 200a are labeled as “agents.”

[0061] In this example, I/O requests from multiple applications (APP1-APP3) running on the host 102 are being processed concurrently over multiple TCP/IP connections. Specifically, each of the applications has placed a respective I/O request, R1-R3, into a request queue 730. When an I/O request is placed into the request queue 730, the next available agent 200a having a connection to the relevant block server responds by retrieving the request from the queue and beginning processing of the request. In this example, a first agent, Agent 1, retrieves request R1 and begins processing of R1 over its respective socket connection. Similarly, request R3 is retrieved by Agent 5, which processes the request to completion over its respective socket. The I/O request R2 from APP2, however involves a larger data transfer. The host driver therefore parses R2 into three constituent I/O requests or operations, each of which is processed by a respective agent (Agents 2, 3 and 4) over a respective socket connection in the same manner as an ordinary I/O request. The sub-requests of I/O request R2 are thus processed in parallel over multiple connections, allowing for increased performance.

[0062] In one embodiment, I/O requests involving data transfers of over 64 kilobytes are subdivided into I/O requests of blocks of 64 k bytes or less. Each constituent I/O request is appropriately tagged with an identifier of the original application-level I/O request to which it corresponds. Once all of the constituent I/O requests are completed, the host driver reports the completion of the application-level I/O request. An important benefit of this feature is that the amount of network bandwidth allocated to each I/O request is generally proportional to the amount of I/O data being transferred.

[0063] III. Allocation of Partitions to Hosts

[0064] One feature that may be incorporated into the architecture is a mechanism for effectively dividing the physical storage space of each block server 104 into smaller units, and allocating or assigning these units to individual hosts 102. Using this feature, a block server can be shared by multiple hosts 102, each of which may “see” its own private disk drive or set of disk drives. One embodiment of this feature is depicted in FIG. 8.

[0065] As illustrated in FIG. 8, the total disk space of each block server 104 may be segmented or subdivided into multiple, variable-size data storage units or “partitions,” each of which may have access privileges independently of other partitions. These storage partitions are preferably created and allocated by a network administrator using either a standard web browser 810 to connect to a configuration and management program 820 running on the block server 104, or by using a command-line interface utility to connect directly to a command shell interpreter running on the block server 104 (not illustrated).

[0066] Each partition can be independently configured with attributes such as size, “read-only,” “read-write,” or zero data before access, and with a list of authorized hosts 102. For example, Partition #1 could be configured to be “20 GB read-only from Host A and read-write from Hosts B and C,” while Partition #2 could be configured as “40 GB read-write from Host A only.” In addition, each partition can be securely exported to a different host computer 102 in a manner which allows the partition to be viewed and managed by the host's user-level software processes as a separate physical disk drive.

[0067] In a preferred embodiment, each partition can include the storage space of multiple disk drives 112. In addition, the storage space of any given disk drive 112 can be subdivided for purposes of creating partitions. For each of the one or more partitions currently defined within the block server 104, the configuration/management program 820 maintains a respective partition table 830 in flash or other non-volatile memory 840 of the block server 104. This table 830 includes configuration data specifying the disk drive(s) and sectors included in the partition, the IP address(es) of the one or more hosts 102 that are authorized to use that partition, and the associated access privileges. The partition tables 830, or another configuration structure stored in the non-volatile memory 840, also preferably includes default or administrator-specified data indicating the maximum number of socket connections that may be established by each host for I/O operations with that block server. As described in the following subsection, during boot-up, a host 102 preferably establishes a dedicated configuration socket connection 400 to each block server 104 it locates, and uses this configuration connection to obtain a list of the partitions that are assigned to it (if any) and the sizes of these partitions.

[0068] In one implementation, up to one hundred partitions can be created per block server 104, and up to sixteen partitions can be assigned to any given host 102. Each host 102 sees each of its partitions starting at logical block address (LBA) zero. The block server 104 maps the (IP address, partition number, LBA) tuple sent in each I/O request into a “physical LBA” for the I/O request. Access to partitions is preferably granted or denied by the block server 104 based on the IP address of the requesting host 102.

[0069] Using this feature of the architecture, network administrators can flexibly and dynamically allocate block server storage resources to hosts 104 transparently to the application-level software running on the hosts. In the example shown in FIG. 8, the block server 104 has five segments or partitions, each having a variable size. Partition 0 represents the unassigned or “free” disk space available on the block server, which may be used to create additional partitions and/or to increase the sizes of previously allocated partitions. Initially, partitions 1, 2 and 3 in this example were created and assigned to hosts 1, 2 and 3, respectively. Subsequently, in order to accommodate an increased storage demand of host 1, a new partition, partition 4, was created and assigned to host 1. The two partitions (1 and 4) uniquely allocated host 1 may be configured to appear as either a single drive segment of a disk drive installed directly in the host 1 computer 102, or as two distinct and separate disk drives. Although the partitions are shown in FIG. 8 as being uniquely assigned to hosts 102, a partition can be assigned to a group of two or more hosts 102.

[0070] The configuration/management program 820 preferably supports a variety of commands and functions for permitting administrators to create and manage partitions. In addition to those functions mentioned above, the program 820 provides functions for deleting partitions and for changing partition ownership. Space from deleted partitions is preferably reclaimed, and re-allocated, using a “best fit” algorithm.

[0071] IV. Authentication and Discovery

[0072] Another feature that may be incorporated into the architecture is a protocol for allowing the block servers 104 to authenticate the hosts 102. Specifically, rather than allowing I/O to proceed immediately, each block server 104 may be designed to authenticate a new host 102 before allowing that host to perform I/O operations. In a preferred embodiment, the authentication process is performed as part of an “authentication and discovery” protocol embodied within the host and server side drivers 204, 206.

[0073] Specifically, in a preferred embodiment, a host 102 locates the available block servers 104 at boot-up time using one of one of two methods: (1) using a host-side configuration file previously set up by an administrator, or (2) using a broadcast packet to discover all of the block servers 104 on its subnet. Each block server 104 preferably includes a daemon process that allows it to be located and configured remotely over the network, but without IP, according to method 2. For each block server 104 located, the host then enters into an authentication and discovery phase, a preferred embodiment of which is illustrated in FIG. 8 without error states.

[0074] As depicted by state 910 in FIG. 9, the host 102 initially sends a connection request to the located block server 104, and the block server 104 responds in state 915 by accepting the request. In state 920, the block server 104 returns a list of software versions that are supported by the block server 104, as well as a random number generated by the block server which serves as an authentication “challenge.” In state 925, the host 102 selects the appropriate version from the list and replies to the block server 104. The host 102 also performs a predefined set of mathematical operations on the random number received from the block server 104. These operations are preferably different for each version of the protocol, making it easy to strengthen the security architecture simply by adding new versions. The number derived from the random “challenge” sent by the block server 104, as the result of the mathematical operations as specified by the version, is sent back to the block server 104 as the host's “response” to the challenge. The block server 104 then authenticates the host 102 in state 930 by determining the expected response (given the challenge sent and the version that was selected), and comparing this expected response to the actual response from the host 102. If the response does not match the expected response, access is denied and all the sockets between the host 102 and the block server 104 are closed (not shown). Otherwise, in state 935, the block server 104 sends an acknowledgement back to the host 102 indicating that the authentication was successful.

[0075] In state 940, the host 102 sends a request to determine the available capacity allocated to it in the given block server 104. In state 945, the block server 104 responds to the host 102 with data specifying the number of partitions assigned to that host and the sizes of these partitions. The block server 104 also returns data specifying the maximum number of data connections that may be established between the host 102 and the block server 104 for I/O traffic, and specifying the dynamic ports that may be used to establish these connections. In state 950, the block server 104 establishes “listen” sockets for traffic originating from an authorized host 102. The host may thereafter establish connections to these sockets up to the maximum number of data connections specified by the block server.

[0076] Thus, upon completion of the authentication and discovery phase between a host 102 and a particular block server 104, the host 102 has been informed of the following information about that block server: (1) which dynamic ports the host can connect to for data, (2) the number of partitions assigned to the host, and (3) the sizes of these partitions.

[0077] Additional details of a commercial implementation of the authentication and discovery phase are provided in Appendix D. The term “Network Storage Unit,” or “NSU,” is used throughout Appendix D to refer to a commercial implementation of the block servers 104.

[0078] Although the invention has been described in terms of certain preferred embodiments, other embodiments that are apparent to those of ordinary skill in the art, including embodiments which do not provide all of the features and advantages set forth herein, are also within the scope of this invention. Accordingly, the scope of the invention is defined by the appended claims.

Claims

1. A block-level shared network storage system, comprising: a storage server comprising an array of disk drives, and comprising a processor that runs a device driver to provide block-level access to data stored on the array of disk drives; and a host computer coupled to the storage server by at least one computer network; wherein the host computer and the storage server perform input/output (I/O) operations over the at least one network using multiple, concurrent logical connections such that a first I/O operation is executed over a first logical connection while a second I/O operation is executed over a second logical connection.

2. The network storage system as in claim 1, wherein each logical connection is a socket connection.

3. The network storage system as in claim 2, wherein each socket connection is a TCP/IP socket connection.

4. The network storage system as in claim 1, wherein each logical connection remains persistent over multiple I/O operations performed over that logical connection.

5. The network storage system as in claim 1, wherein the host computer is programmed to divide an I/O operation into multiple constituent I/O operations, and to perform the multiple constituent I/O operations in parallel over respective logical connections of said multiple, concurrent logical connections.

6. The network storage system as in claim 1, wherein the storage server provides virtualized block-level storage access to the host computer such that the block server appears as local disk drive storage to user-level processes running on the host computer.

7. The network storage system as in claim 1, wherein the storage server is configurable to provide multiple storage partitions, each of which may be allocated to a different host computer.

8. The network storage system as in claim 1, wherein the storage server has a first storage partition which is uniquely assigned to the host computer such that the first storage partition appears to user-level processes running on the host computer as a private, local disk drive.

9. The network storage system as in claim 8, wherein the storage server further has a second storage partition which is uniquely assigned to a second host computer.

10. The network storage system as in claim 8, wherein the storage server further has a second storage partition which is shared by multiple host computers.

11. The network storage system as in claim 1, wherein the host computer and the storage server implement an authentication protocol in which the storage server authenticates the host computer before allowing the host computer to perform input/output operations.

12. The network storage system as in claim 1, wherein the host computer and the storage server implement a discovery protocol in which the storage server notifies the host computer of partitions assigned to the host computer.

13. The network storage system as in claim 1, wherein at least one of the logical connections is over a general-purpose computer network.

14. The network storage system as in claim 1, wherein at least one of the logical connections is over an Ethernet network.

15. The network storage system as in claim 1, wherein the first and second logical connections exist over separate computer networks.

16. The network storage system as in claim 1, wherein each logical connection exists between a respective reader/writer pair.

17. The network storage system as in claim 1, wherein the host computer and the storage server are interconnected by at least one switch.

18. The network storage system as in claim 1, wherein the host computer and the storage server each include two network interfaces that provide redundant network connections between the host computer and the storage server.

19. A system for storing data for host computers, comprising: a plurality of storage servers connected to a network, each storage server comprising an array of disk drives, an array controller, and a processor; a plurality of host computers connected to the network and programmed to store data on the storage servers; and at least one switch which interconnects the plurality of storage servers with the plurality of host computers; wherein each host computer is programmed to open multiple concurrent sockets over the network to the storage servers for performing concurrent input/output operations.

20. The system of claim 19, wherein the sockets are TCP/IP sockets.

21. The system of claim 19, wherein each storage server of the plurality of storage servers provides virtualized block-level storage access to the host computers such that the block servers appear as local disk drive storage to user-level processes running on the host computers.

22. The system of claim 19, wherein at least a first host computer of the plurality of host computers is programmed to divide an I/O operation into multiple constituent I/O operations, and to perform the multiple constituent I/O operations in parallel over respective logical socket connections between the first host computer and a target storage server.

23. The system of claim 19, wherein a first storage server of the plurality of storage servers is configurable to provide multiple, variable-size partitions, each of which may be allocated to a different host computer of the plurality of host computers.

24. The system of claim 19, wherein a first storage server of the plurality of storage servers has a first partition which is uniquely assigned to a first host computer of the plurality of host computers such that the first partition appears as a local disk drive to the first host computer.

25. The system of claim 24, wherein the first storage server further has a second partition which is uniquely assigned to a second host computer of the plurality of host computers.

26. The system of claim 19, wherein the host computers and the storage servers implement an authentication protocol in which a storage server authenticates a host computer before allowing the host computer to perform input/output operations.

27. The system of claim 19, wherein the host computers and the storage servers implement a discovery protocol in a storage server notifies a host computer of partitions assigned to the host computer.

28. A method of performing input/output operations, comprising: establishing first and second TCP/IP connections between a host computer and a block-level storage server over one or more computer networks; performing a first input/output operation over the first TCP/IP connection while performing a second input/output operation over the second TCP/IP connection, each of said input/output operations comprising a transfer of input/output data between the host computer and the storage server; and maintaining the first and second TCP/IP connections is a persistent state such that each TCP/IP connection may be used to perform additional input/output operations.

29. The method as in claim 28, wherein the first and second TCP/IP connections are established over separate computer networks.

30. The method as in claim 28, wherein the first input/output operation is a first I/O request issued from a first application running on the host computer, and the second input/output operation is a second I/O request issued from a second application running on the host computer.

31. The method as in claim 28, wherein the first and second input/output operations are constituent operations of an I/O request issued by a process running on the host computer, whereby the I/O request is executed in parallel over multiple TCP/IP connections.

32. The method as in claim 28, further comprising establishing a third TCP/IP connection between the host computer and the storage server, and using the third TCP/IP connection to perform an authentication sequence in which the storage server authenticates the host computer.

33. The method as in claim 32, further comprising conveying access information over the third TCP/IP connection from the storage server to the host computer, said access information specifying access rights uniquely assigned to the host computer.

34. A method of executing an input/output (I/O) request received from a user-level process running on a host computer, comprising: on the host computer, dividing the I/O request into multiple constituent I/O operations; and performing the multiple constituent I/O operations in parallel over multiple, respective logical network connections between the host computer and a target storage server such that I/O data is transferred between the host computer and the storage server over each of the logical network connections.

35. The method of claim 34, wherein each logical network connection is a socket connection.

36. The method of claim 34, wherein each logical network connection is a TCP/IP socket connection.

37. The method of claim 34, wherein at least one of the logical network connections is over a general-purpose computer network.

38. A storage server system that provides block-level storage for host computers over a computer network, the system comprising: a storage server comprising a processor, an array of disk drives coupled to a controller, and a network interface for connecting the storage server to a computer network; and a software system which provides functionality for subdividing storage space of the array of disk drives into multiple storage partitions, and provides functionality for uniquely allocating the partitions to host computers on the network such that the storage server may be shared by multiple host computers, and such that when a partition is allocated to a host computer, the partition appears to user-level processes of the host computer as a local disk drive.

39. The storage server system as in claim 38, wherein the software system comprises a host driver component which runs on the host computers, and comprises a server driver component executed by the processor of the storage server, wherein the host and server driver components communicate over the computer network via TCP/IP connections.

40. The storage server system of claim 38, wherein the software system further comprises a configuration and management program which runs on the storage server and provides functionality for partitions to be created and assigned to hosts over the computer network using a web browser.

41. The storage server system of claim 38, wherein the software system further provides functionality for allocating a partition to multiple host computers to permit sharing of partitions.

42. The storage server system of claim 38, wherein the partitions have a user-definable size.

43. The storage server system of claim 38, wherein the software system supports the ability for a host computer to concurrently perform multiple input/operations over multiple, respective TCP/IP connections to the storage server.

44. The storage server system of claim 38, wherein the software system implements an authentication protocol in which the storage server authenticates host computers using a challenge-response protocol.

45. The storage server system of claim 38, wherein the software system implements a discovery protocol in which the storage server notifies the host computer of storage partitions assigned to the host computer.

46. A method of providing storage access over a computer network, comprising: establishing a logical connection over a computer network between a host computer and a storage server; performing an authentication sequence over the logical connection whereby the storage server authenticates the host computer; when the authentication sequence fails, terminating the logical connection and denying the host computer access to storage resources of the storage server; and when the authentication sequence is successful, providing access information over the logical connection from the storage server to the host computer to allow the host computer to access the storage resources, at least some of the access information being unique to the host computer.

47. The method of claim 46, wherein the logical connection is a TCP/IP connection.

48. The method of claim 46, wherein the authentication sequence comprises the storage server transmitting a randomized challenge to the host computer and thereafter determining whether a host response to the randomized challenge matches an expected response.

49. The method of claim 46, wherein the authentication sequence further comprises selecting a software version which is common to the storage server and the host computer, and selecting an authentication method which is dependent upon the selected software version.

50. The method of claim 46, wherein the access information specifies a set of dynamic ports to which the host computer may connect to perform input/output operations.

51. The method of claim 46, wherein the access information specifies a maximum number of logical connections that may be established by the host computer to the storage server.

52. The method of claim 46, wherein the access information specifies a number of storage partitions allocated to the host computer and sizes of such storage partitions.