US20110072104A2 - Pull data replication model - Google Patents

Abstract

A data storage system, device, and method are provided for replicating data between different data storage systems or appliances. More specifically, the present invention allows an initiating data storage device to pull data from a target data storage device. Thus, the storage device that is supplying the data can act as a target and does not require the additional software inherent in being an initiator.

Description

CROSS REFERENCE TO RELATED APPLICATIONS
• This application claims the benefit of U.S. Provisional Application No. 60/771,384, filed Feb. 7, 2006, the entire disclosure of which is hereby incorporated herein by reference. Also, this application is related to U.S. patent application Ser. No. 11/561,512, filed Nov. 20, 2006, entitled Data Replication Method and Apparatus.
FIELD
• The present invention is directed to data replication. In particular, the present invention is directed to methods and apparatuses for replicating data between different data storage systems or appliances.
BACKGROUND
• The need to store digital files, documents, pictures, images and other data continues to increase rapidly. In connection with the electronic storage of data, various data storage systems have been devised for the rapid and secure storage of large amounts of data. Such systems may include one or a plurality of storage devices that are used in a coordinated fashion. Systems in which data can be distributed across multiple storage devices such that data will not be irretrievably lost if one of the storage devices (or in some cases, more than one storage device) fails are also available. Systems that coordinate operation of a number of individual storage devices can also provide improved data access and/or storage times. Examples of systems that can provide such advantages can be found in the various RAID (redundant array of independent disks) levels that have been developed. Whether implemented using one or a plurality of storage devices, the storage provided by a data storage system can be treated as one or more storage volumes.
• In order to facilitate the availability of desired data, it is often desirable to maintain different versions of a data storage volume. Indeed, data storage systems are available that can provide at least limited data archiving through backup facilities and/or snapshot facilities. These facilities can comprise automated or semi-automated batch replication facilities. By maintaining different versions, disaster recovery is facilitated. For example, if a virus causes a current storage volume version to be lost or otherwise unusable, the system can be rolled back to an earlier version that does not include the file that introduced the virus. In order to further facilitate the use and security of data, it is often desirable to create a copy of a data storage volume that is originally maintained by a first data storage system on a second data storage system. However, creating backup copies or snapshots of a first data storage system on a second data storage system, commonly referred to as remote replication, introduces complications. In particular, a data storage system that is the source of the data that is to be replicated must act as an initiator device in order to move the data to the recipient data storage system that is acting as the target. However, data storage systems typically are targets, not initiators. Furthermore, an initiator is a more complex and difficult to implement function, requiring more intelligence and processing power to implement as compared to a typical target device. Accordingly, in order for a source data storage system to perform data replication in cooperation with a remote data storage system, the source data storage system must typically be enhanced to include initiator functions, as well as remote replication functions.
• Data replication between different storage systems is further complicated where the initiator and the target data storage systems are from different data storage system vendors. This is because different vendors typically use different protocols and instructions to control operation of their data storage systems. As a result, the automated or semi-automated batch replication of data between data storage systems from different vendors is at best difficult, and is often impossible.
SUMMARY
• The present invention is directed to solving these and other problems and disadvantages of the prior art. In accordance with embodiments of the present invention, a remote batch data replication service is provided that consists of asynchronous replication of data between a local and a remote system through the use of snapshots. In one embodiment, the remote batch data replication service features asynchronous replication of block-level data from a volume at a local data storage system to a remote data storage system. More particularly, in a pull data replication model, the local or source data storage system is able to operate as a target device. In accordance with still other embodiments of the present invention, the source and remote data storage systems can exchange information regarding their respective capabilities, and can use a common protocol to enable data replication even where the data storage systems are from different vendors (i.e. they are heterogeneous systems).
• In order to allow a source data storage system to operate as a target device while replicating data to a remote or destination data storage system, the destination data storage system operates to pull data from the source data storage system. Pulling data can include making use of data snapshot capabilities native to the source data storage system. In accordance with embodiments of the present invention, the destination data storage system commands the source data storage system to take a first snapshot of the storage volume that is to be replicated. The source data storage system is also instructed to make the first snapshot visible to the destination data storage system, and the destination data storage system makes a copy of the data contained in the first snapshot. That is, the destination data storage system copies the complete data storage volume. The destination data storage system copies all of the blocks from the first snapshot and then the destination takes a snapshot. The destination data storage system thereafter commands the source storage system to take another snapshot, and requests a list of the data block numbers that have changed. A copy of the changed data blocks is then requested from the source data storage system by the destination data storage system. By copying only the changed data blocks, a complete backup copy of the replicated storage volume can be maintained on the destination data storage system, without requiring copying the complete storage volume at each update of the backup copy. Furthermore, embodiments of the present invention do not require that the source data storage system have the additional intelligence and features required to operate as an initiator. Instead, the source data storage system can operate as a target by responding to commands initiated by the destination data storage system, or by some other initiator.
• In accordance with further embodiments of the present invention, data replication and other features can be performed between heterogeneous data storage systems using a heterogeneous communication protocol. The heterogeneous communication protocol provides a bridge between data storage systems from different vendors, and allows heterogeneous data storage systems to advertise and discover their respective data replication capabilities. In order to implement the heterogeneous communication or bridging protocol, the vendors of different data storage systems each use a common replication protocol. In accordance with embodiments of the present invention, in response to a query by a remote data storage system using the heterogeneous communication protocol, a source data storage system can be instructed to advertise or otherwise provide information regarding its capabilities to the remote data storage system. It should be noted that the remote data storage system could also be commanded to provide information regarding its capabilities. Furthermore, data replication between heterogeneous data storage systems can be performed with the source data storage system operating as a target. Although embodiments implementing a heterogeneous communication protocol allow replication of data from a source data storage system operating as a target device to a remote data storage system operating as an initiator, it should be appreciated that the heterogeneous communication protocol can support other types of communications between heterogeneous data storage systems. For example, the heterogeneous replication protocol provides for data replication in either direction between two storage systems from the same or different vendors.
• In one embodiment, a simple heterogeneous replication protocol is a simple remote data replication protocol that provides the capability for systems that do not inherently implement remote data replication to perform remote data replication with another system as both a source and target of replication data. Simple heterogeneous replication protocol may be implemented as a cross-vendor/cross platform protocol. The system that does not inherently implement remote data replication implements snapshot capability and provides the ability to map a newly created snapshot to a Logical Unit Number (LUN) or other type of mechanism defined in the protocol capable of getting the data. The system also implements simple heterogeneous replication protocol such that a partner system can query data management services capabilities (e.g., SHRP, snapshot, split mirror, replication, etc.) and issue commands to control the snapshot services in order to facilitate asynchronous batch (snapshot based) replication.
• Additional features and advantages of embodiments of the present invention will become more readily apparent from the following description, particularly when taken together with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a functional block diagram depicting components of an electronic data system incorporating a source data storage system and a remote data storage system in accordance with embodiments of the present invention;
• FIG. 2A is a block diagram depicting components of a data storage system in accordance with embodiments of the present invention;
• FIG. 2B is a block diagram depicting components of a data storage system in accordance with other embodiments of the present invention;
• FIG. 3A is a block diagram depicting components of a storage appliance in accordance with embodiments of the present invention;
• FIG. 3B is a block diagram depicting components of a storage controller in accordance with embodiments of the present invention;
• FIG. 4 is a flow chart depicting aspects of a data replication process in accordance with embodiments of the present invention;
• FIG. 5 is a flow chart depicting aspects of a process for exchanging information between heterogeneous data storage systems in accordance with embodiments of the present invention; and
• FIG. 6 depicts an exemplary data structure of a response to a destination storage system's query command in accordance with embodiments of the present invention.
DETAILED DESCRIPTION
• In accordance with embodiments of the present invention, a snapshot is a block level point-in-time representation of data on a storage volume. The data is essentially frozen in time at the instant that the snapshot is taken. Although data on the storage volume may change as a result of write operations, the data represented by the snapshot will remain constant and frozen in time at the instant that the snapshot was taken. In order to preserve snapshot data, a repository is used to store data that is not otherwise represented in the storage volume and snapshot metadata. All data and metadata associated with the snapshot is stored in the repository, although such storage on the repository is not required. In accordance with embodiments of the present invention, data stored within the snapshot is stored in “chunks.” A chunk is equivalent to a number of logical data blocks (LBAs). As a further optimization, data may also be stored at a subchunk level. A subchunk is a fixed size subset of a chunk. Accordingly, data can be moved betweendata storage systems104,108 in units of chunks, subchunks, or any multiple thereof. The units that are used for data replication operations can be selected to optimize the performance of the network link between the data storage systems.
• FIG. 1 is a block diagram depicting anelectronic data system100 in accordance with embodiments of the present invention incorporating a firstdata storage system104 and a seconddata storage system108. Theelectronic data system100 may also include one or more host processors, computers orcomputer systems112. In addition, theelectronic data system100 may include or may be interconnected to anadministrative computer116. As will be appreciated by one of skill in the art after consideration of the present disclosure, embodiments of the present invention have applications in association with single ormultiple hosts112 in storage area network (SAN) or direct connect environments.
• Thedata storage systems104,108 are typically interconnected to one another through an in-band network120. The in-band network120 may also interconnect thedata storage systems104,108 to ahost computer112 and/or anadministrative computer116. Theelectronic data system100 may also include an out-of-band network124 interconnecting some or all of theelectronic data system100nodes104,108,112 and/or116. For example, in a typical remote replication environment, one ormore host computers112 are connected to eachdata storage system104,108. For instance, a firstdata storage system104 is connected to a seconddata storage system108 across some distance by a Fibre Channel or a TCP/IP network120, and each of thesedata storage systems104,108 is connected to ahost computer112 through an in-band120 and/or an out-of-band124 network.
• The in-band orstorage area network120 generally functions to transport data betweendata storage systems104 and/or108 andhost devices112, and can be any data pipe capable of supporting multiple initiators and targets. Accordingly, examples of in-band networks120 include Fibre Channel (FC), iSCSI, parallel SCSI, Ethernet, ESCON, or FICON connections or networks, which may typically be characterized by an ability to transfer relatively large amounts of data at medium to high bandwidths. The out-of-band network124 generally functions to support the transfer of communications and/or commands between various network nodes, such as datastorage resource systems104,108,host devices112, and/oradministrative computers116, although such data may also be transferred over the in-band communication network120. Examples of an out-of-band communication network124 include a local area network (LAN) or other transmission control protocol/Internet protocol (TCP/IP) network. In general, the out-of-band communication network124 is characterized by an ability to interconnect disparate nodes or other devices through uniform user interfaces, such as a web browser. Furthermore, the out-of-band communication network may provide the potential for globally or other widely distributed management or globally distributed remote replication betweendata storage systems104,108 via TCP/IP.
• Every electronic data system node orcomputer104,108,112 and116, need not be interconnected to every other node or device through both the in-band network120 and the out-of-band network124. For example, nohost device112 needs to be interconnected to anyother host device112,data storage system104,108, oradministrative computer116 through the out-of-band communication network124, although interconnections between ahost device112 andother devices104,108,116 through the out-of-band communication network124 are not prohibited. As another example, anadministrative computer116 may be interconnected to at least onestorage resource device104 or108 through the out-of-band communication network124. Anadministrator computer116 may also be interconnected to the in-band network120 directly, although such an interconnection is not required. For example, instead of a direct connection, anadministrator computer116 may communicate using the in-band network120 to a controller of adata storage system104,108.
• In general, ahost computer112 exchanges data with one or more of thedata storage systems104,108 in connection with the performance of the execution of application programming, whether that application programming concerns data management or otherwise. Furthermore, anelectronic data system100 may includemultiple host computers112. Anadministrative computer116 may provide a user interface for controlling aspects of the operation of thestorage systems104,108. Theadministrative computer116 may be interconnected to thestorage system104 directly, and/or through a bus ornetwork120 and/or124. In accordance with still other embodiments of the present invention, anadministrative computer116 may be integrated with ahost computer112. In addition, multipleadministrative computers116 may be provided as part of theelectronic data system100. Furthermore, although only twodata storage systems104,108 are shown inFIG. 1, anelectronic data system100 may include more than two data storage systems.
• FIG. 2A illustrates components that may be included in adata storage system104,108 in accordance with embodiments of the present invention. In general, thedata storage system104,108 includes a number ofstorage devices204. Examples ofstorage devices204 include hard disk drives, such as serial advanced technology attachment (SATA), small computer system interface (SCSI), serial attached SCSI (SAS), Fibre Channel (FC), or parallel advanced technology attached (PATA) hard disk drives. Other examples ofstorage devices204 include magnetic tape storage devices, optical storage devices or solid state disk devices. Furthermore, although a number ofstorage devices204 are illustrated, it should be appreciated that embodiments of the present invention are not limited to any particular number of storage devices, and that a lesser or greater number ofstorage devices204 may be provided as part of adata storage system104. As can be appreciated by one of skill in the art, one or more arrays and/or array partitions, hereinafter referred to as logical unit numbers (LUNs) comprising a storage volume, may be established on thedata storage devices204. As used herein, an array is understood to refer to a RAID set and the array partition. It can be appreciated that some subset of the array partition can be represented as a single unit to a host. As can be further appreciated by one of skill in the art, a LUN may be implemented in accordance with any one of the various array levels or other arrangements for storing data on one ormore storage devices104. As can also be appreciated by one of skill in the art, thestorage devices204 may contain data comprising a master storage volume, which may correspond to a LUN, in addition to one or more snapshots of the master storage volume taken at different times. As can further be appreciated by one of skill in the art, snapshots may comprise metadata and data stored in a backing store on thestorage devices204. As can also be appreciated by one of skill in the art, thestorage devices204 contain data comprising a master storage volume, which may correspond to a LUN, and one or more snapshots. In one embodiment, the snapshots may be mapped to the LUNs.
• Adata storage system104,108 in accordance with embodiments of the present invention may be provided with afirst controller slot208a. In addition, other embodiments may include additional controller slots, such as asecond controller slot208b. As can be appreciated by one of skill in the art, acontroller slot208 may comprise a connection or set of connections to enable acontroller212 to be operably interconnected to other components of thedata storage system104,108. Furthermore, adata storage system104,108 in accordance with embodiments of the present invention includes at least onecontroller212a. For example, while thedata storage system104,108 is operated in a single controller, non-failover mode, thedata storage system104,108 may include exactly onecontroller212. Adata storage system104,108 in accordance with other embodiments of the present invention may be operated in a dual redundant active-active controller mode by providing asecond controller212b. When asecond controller212bis used in addition to afirst controller212a, thesecond controller slot208breceives the second controller. As can be appreciated by one of skill in the art, the provision of two controllers,212ato212b, permits data to be mirrored between thecontrollers212a-212b, providing redundant active-active controller operation.
• One or more busses orchannels216 are generally provided to interconnect a controller orcontrollers212 through the associated controller slot orslots208 to thestorage devices204. Furthermore, while illustrated as a single shared bus orchannel216, it can be appreciated that a number of dedicated and/or shared buses or channels may be provided. Additional components that may be included in adata storage system104 include one ormore power supplies128 and one ormore cooling units132. In addition, a bus ornetwork interface136 may be provided to interconnect thedata storage system104,108 to the bus ornetwork112, and/or to ahost computer108 oradministrative computer116.
• Although illustrated as a complete RAID system inFIG. 2B, it should be appreciated that thedata storage system104 can comprise one or more storage volumes implemented in various other ways. For example, thedata storage system104 may comprise a hard disk drive orother storage device204 connected or associated with a server or a general purpose computer. As further examples, thestorage system104 may comprise a Just a Bunch of Disks (JBOD) system or a Switched Bunch of Disks (SBOD) system.
• With reference toFIG. 2B, components that may be included in adata storage system104,108 in accordance with other embodiments of the present invention are illustrated. In general, thedata storage system104,108 according to such embodiments includes astorage appliance220 interconnecting one ormore storage devices204 to a bus or network. Furthermore, thestorage appliance220 may be inserted between ahost112 or other device and one ormore storage devices204. In accordance with still other embodiments of the present invention, astorage device204 may itself comprise a collection of hard disk drives or other storage devices, for example provided as part of a RAID or SBOD system.
• With reference toFIG. 3A, astorage appliance220 in connection with an embodiment of the present invention in which at least some data replication establishment and management functions are provided by software running on thestorage appliance220 interconnecting one ormore storage devices204 to ahost device112 or otherdata storage systems104,108 is illustrated. The components may include aprocessor304acapable of executing program instructions. Accordingly, theprocessor304amay include any general purpose programmable processor or controller for executing application programming. Alternatively, theprocessor304amay comprise a specially configured application specific integrated circuit (ASIC). Theprocessor304agenerally functions to run programming code including adata replication application328a.
• Thehost112 or the administrative116 computer may additionally include memory308afor use in connection with data replication applications. For example, the memory308amay store a copy of thedata replication application328aconfiguration instructions. The memory308amay comprise solid state memory resident, removable or remote in nature, such as FLASH, DRAM and/or SDRAM.
• Thehost112 or the administrative116 computer may also includedata storage314afor the storage of application programming and or data. For example,operating system software318 may be stored in thedata storage314a. In addition, thedata storage314amay be used to store adata replication application328acomprising instructions for pulling data from a sourcedata storage system104 and providing the data to adestination storage volume108 as described herein. Thedata replication application328amay itself include a number of modules or components, such as a main input/output (IO) module and a restore thread or module.
• Thehost112 or the administrative116 computer may also include one ormore network interfaces332a. For example, afirst network interface332a1 may interconnect the storage appliance to ahost device112 through a first network and asecond network interface332amay interconnect the storage appliances to the storage device ordevices204 that, together with thestorage appliance220, comprise adata storage system104,108. Furthermore, the first and second networks may be of the same type or different types, and the storage appliance200 may be “inline” between ahost device112 and astorage device204. Examples of anetwork interface332ainclude a fiber channel (FC) interface, Ethernet, or any other type of communication interface. As can be appreciated by one of skill in the art, anetwork interface332amay be provided in the form of a network interface card or other adapter.
• FIG. 3B illustrates aspects of astorage controller212 in accordance with embodiments of the present invention. In general, astorage controller212 includes aprocessor subsystem304 capable of executing instructions for performing, implementing and or controllingvarious controller212 functions. Such instructions may include instructions for implementing aspects of a snapshot restore method and apparatus. Furthermore, such instructions may be stored as software and/or firmware. As can be appreciated by one of skill in the art, operations concerning the generation of parity data or other operations may be performed using one or more hardwired and or programmable logic circuits provided as part of theprocessor subsystem304. Accordingly, theprocessor subsystem304 may be implemented as a number of discrete components, such as one or more programmable processors in combination with one or more logic circuits.Processor subsystem304 may also include or be implemented as one or more integrated devices or processors. For example a processor subsystem may comprise a complex programmable logic device (LPLD).
• Acontroller212 also generally includesmemory306. Thememory306 is not specifically limited to memory of any particular type. For example, thememory306 may comprise a solid state memory device, or a number of solid state memory devices. In addition, thememory306 may include separatevolatile memory308 andnon-volatile memory310 portions. As can be appreciated by one of skill in the art, thememory306 typically includes awrite cache312 and aread cache316 that are provided as part of thevolatile memory308 portion of thememory306, although other arrangements are possible. By providingcaches312,316, astorage controller212 can improve the speed of input/output (IO) operations between ahost108 and thedata storage devices204 comprising an array or array partition. Examples ofvolatile memory308 include DRAM and SDRAM.
• Thenon-volatile memory310 may be used to store data that was written to thewrite cache312 ofmemory306 in the event of a power outage affecting thedata storage system104. Thenon-volatile memory portion310 of thestorage controller memory306 may include any type of data memory device that is capable of retaining data without requiring power from an external source. Examples ofnon-volatile memory310 include, but are not limited to, compact flash or other standardized non-volatile memory devices.
• Thememory306 also includes portions of thememory306 comprising aregion324 that provides storage forcontroller code326. Thecontroller code326 may comprise a number of components, including a data replication process orapplication328 comprising instructions for pulling data from a source data storage system as described herein. Thedata replication application328 may itself include a number of modules or components, such as a main input/output (IO) module and a restore thread or module. As shown inFIG. 3B, thecontroller code region324 may be established in avolatile memory308 portion of thestorage controller memory306. Alternatively or in addition,controller code326 may be stored innon-volatile memory310.
• Astorage controller212 may additionally include other components. For example, a bus and/ornetwork interface332 may be provided for operably interconnecting thestorage controller212 to the remainder of thedata storage system104, for example through acontroller slot208 and a bus orchannel216. Furthermore, theinterface332 may be configured to facilitate removal or replacement of thestorage controller212 in acontroller slot208 as a field replaceable unit (FRU). In addition, integral signal and power channels may be provided for interconnecting the various components of thestorage controller212 to one another.
• With reference toFIG. 4, a process for replicating data stored on a sourcedata storage system104 to a destinationdata storage system108, in which the sourcedata storage system104 has the attributes of a target device, are illustrated. Initially, atstep404, the destinationdata storage system108 or a storage appliance (hereinafter referred to as the initiator) commands the sourcedata storage system104 to take a snapshot of the master storage volume maintained by the source data storage system104 (or of some specified storage volume maintained by the source data storage system104). As used herein, a snapshot is a virtual volume that represents the data that existed on the master storage volume at the point in time that the snapshot was taken. The master storage volume is the current set of data maintained on the sourcedata storage system104. The master storage volume may correspond to a standard RAID volume or LUN. The initiator then commands the sourcedata storage system104 to map the snapshot to a LUN and to make the LUN visible to the initiator (step408). All of the blocks of data in the snapshot are then read by the destination data storage system108 (or by another device acting as the initiator), and the data is copied to the destination data storage system (e.g., the second data storage system108) (step412). As can be appreciated by one of skill in the art after consideration of the present disclosure, copying the data to thedestination storage system108 generally comprises storing the data indata storage devices204 that are local to or associated with the destinationdata storage system108.
• In accordance with at least one embodiment of the present invention, rather than transferring all data blocks from the sourcedata storage system104 to the destinationdata storage system108, it may be possible to perform a hash operation on a range of blocks on the sourcedata storage system104 and the destinationdata storage system108 in order to create as hash list. The hash list can then be transferred from the sourcedata storage system104 to the destinationdata storage system108. If the hash key for a range of disk blocks on the sourcedata storage system104 is identical to the hash key for the same range of blocks on the destinationdata storage system108, then it is not necessary to transfer the data blocks. The hash algorithm facilitates effective use of the bandwidth of the asynchronous link between the local and remote systems.
• Atstep416, the destinationdata storage system108 takes a snapshot of the data copied from the sourcedata storage system104. Both data storage systems therefore have identical copies of data and a snapshot of the data from the same point in time after this step is completed. It should be noted that the same algorithm can be used to copy any volume, even if not snapped, so long as the source volume is not being written to during the copy. This results in a destination initiated volume copy.
• Atstep420, the initiator commands the sourcedata storage system104 to take another snapshot of the master storage volume. The initiator then commands the sourcedata storage system104 to map this snapshot to a LUN and to make the LUN visible to the destination data storage system108 (step424).
• The initiator next commands the sourcedata storage system104 to provide a list of the changes (i.e. the delta data) between the current snapshot and the previous snapshot (step428). Using the provided list, the initiator then reads the blocks in the list from the source system, and copies the data to the local data storage (step432). The destinationdata storage system108 then takes a snapshot of the data now stored in the local data storage (step436). Thesource104 anddestination108 data storage systems thus have identical snapshots or copies of the master data storage volume on the sourcedata storage system104, at the time the first snapshot was taken, and at the time the second snapshot was taken. If desired, earlier snapshots can be deleted from one or both of thedata storage systems104,108, for example if it is considered unimportant to maintain a number of restore points for the master storage volume.
• Atstep440, a determination can be made as to whether the copy of the master storage volume on thedestination storage device108 should be updated. If it is determined that an update should be performed, the process may return to step420. If no further updates are required, the process may end. In accordance with embodiments of the present invention, copies of the master storage volume are made periodically. For example, an updated copy of the master storage volume can be taken every 10 minutes.
• As can be appreciated by one of skill in the art after consideration of the present disclosure, data can be replicated from a sourcedata storage system104, even if the sourcedata storage system104 does not provide native support for data replication to anotherdata storage system108. In addition, the sourcedata storage system104 does not need to act as an initiator in connection with replicating stored data. Instead, thesource data system104 can operate as a target device. In accordance with embodiments of the present invention where the sourcedata storage system104 is not an initiator, in order to provide data replication as described, the sourcedata storage system104 should be able to respond to commands to take snapshots, to map a snapshot to a LUN, and provide lists of LBAs that contain data (or changed data for subsequent copies). Furthermore, it can be appreciated that, following the transfer of the initial snapshot data comprising a complete copy of the data in the source's master storage volume, only the changed data needs to be transferred in order to establish an updated copy of the master storage volume on the destination storage device.
• These commands necessary to perform data replication may be communicated using a simple communication protocol. In accordance with still other embodiments of the present invention, a simple bridging communication protocol may be provided that enables these and other functions to be performed betweendata storage systems104,108 from different vendors (i.e. between heterogeneous data storage systems). Accordingly, the simple bridging protocol may comprise a set of commands recognized bydata storage system104,108 sourced from different vendors. The simple bridging protocol may be provided to different vendors, in order to facilitate communications necessary in order to replicate data or performing other functions or services, for example between heterogeneous data storage systems. As a further example, a destinationdata storage system108 may communicate with asource communication system104 that natively supports a different native communication protocol, in order to perform data replication as described herein. Therefore, a sourcedata storage system104 from a different vendor than the destinationdata storage system108 need not be provided with any extra intelligence in order to provide data replication in cooperation with the destinationdata storage system108, other than an ability to take snapshots and respond to commands expressed in the simple bridging protocol.
• As can be appreciated by one of skill in the art after consideration of the present disclosure, the simple bridging protocol is not limited to use in connection with pulling data from a sourcedata storage system104 operating as a target. For example, the simple bridging protocol can be used to allow a sourcedata storage system104 to act as an initiator and the destinationdata storage system108 to act as a target in connection with data replication operations. For instance, using the simple bridging protocol, the source data storage system104 (operating as an initiator) replicates data to a volume on thedestination storage system108 and commands the destination storage system108 (operating as a target) to take a snapshot of that volume.
• In accordance with embodiments of the present invention, and with reference toFIG. 5, a process for providing and using a simple bridging protocol comprises loading the protocol into a sourcedata storage system104 and into a destination storage system108 (step504). The protocol may be loaded onto thestorage systems104,108 as a software plug-in or may already be resident within firmware of thedata storage systems104,108. A query for replication capabilities is then issued by thedata storage system104 or108 operating as an initiator to thedata storage systems104 or108 operating as a target (step508). Alternatively, the query may be issued by another initiator in theelectronic data system100. This query may be issued using the simple bridging protocol.
• In response to the query, thedata storage system104 or108 operating as a target device provides capability information (step512). The capability information may comprisedata storage system104 or108 data management services capabilities, such as snapshot capabilities, the ability to act as an initiator, remote data replication capabilities, storage volume naming capabilities, snapshot volume naming capabilities, storage volume mapping capabilities, snapshot volume mapping capabilities, etc. Thedata storage system104 or108 acting as the initiator, or some other initiator in theelectronic data system100, then uses the capability information to determine how to effect data replication between the sourcedata storage system104 and the destination data storage system108 (step516).
• By providing or advertising capability information in response to a query, a system using the simple bridging protocol is not limited to any fixed set of features. Accordingly, as features are added to versions of adata storage system104 or108, those added features may be reported to thedata storage systems104 or108 or other components of an electronicdata storage system100 and can be made available to those other components.
• FIG. 6 depicts an exemplary data structure of a response to a destination storage system's108 query command in accordance with embodiments of the present invention. The response to a query command generally comprises, without limitation,snapshot capability information604, remote datareplication capability information608, storage volumenaming capability information612, snapshot volumenaming capability information616, and other volume, snapshot and/or storage systemcharacteristic information620. When the destinationdata storage system108 receives the information describing the source data storage system's104 capabilities, the destinationdata storage system108 can determine what types and methods of remote data replication can be employed to replicate data to/from the sourcedata storage system104.
• Thesnapshot capability information604 may include the number of supported snapshots for the entire storage system and/or on a per volume basis and how those snapshots are characterized. For example some snapshots may be fully allocated whereas other snapshots may be sparse snapshots and others may be both. Additionally, thesnapshot capability information604 may include snapshot naming formats for snapshots on the system.
• Information included as a part of the remotedata replication capabilities608 may include, for example, whether remote replication is supported on the sourcedata storage system104, and if remote replication is supported what types of remote replication are supported (e.g., transactional asynchronous, batch asynchronous, synchronous, CDP, or any other type of data replication protocol known in the art). The remote datareplication capability information608 may further indicate whether the replying system is capably of acting as a remote replication target and/or as an initiator. This essentially indicates whether the replying system is enabled to operate in a target mode and/or an initiator mode. Additionally, the remote datareplication capability information608 may indicate what type of support the replying system has for a pull data replication model (if any). Also, the number of remote targets supported per storage volume and if remote replication chaining is supported may be included in the remote datareplication capability information608.
• Storagevolume naming capabilities612 may include the protocol used to support volume naming, the maximum number of characters allowed in a storage volume name, the minimum number of characters allowed in a storage volume name, and so on. By providing storagevolume naming capabilities612, the destinationdata storage system108 can name storage volumes as they would be named in the sourcedata storage system104, or at least be aware of the naming schemes in the sourcedata storage system104.
• Similar to storagevolume naming capabilities612, the replying system may identify how snapshots are named by providing snapshotvolume naming capabilities616. The snapshotvolume naming capabilities616 may include, for instance, the maximum number of characters in a snapshot volume name, the minimum number of characters in a snapshot volume name, and any off limits characters in a snapshot volume name.
• As can be appreciated by one of skill in the art, additional information may be provided by the replying system in the other volume, snapshot, and/or storage systemcharacteristics information field620. Data stored in this particular field may include, without limitation, storage system configuration information, the number of storage volumes in use, and any state information related to the storage system, volume, and/or snapshots.
• As can be appreciated, any reliable communication protocol may be used as the bridging protocol to provide communication capabilities betweenstorage systems104,108. The bridge communication protocol used between thestorage systems104,108 in some embodiments may include Fibre Channel and/or iSCSI. The use of iSCSI as the bridge protocol may afford the use of SCSI Command Descriptor Blocks (CDBs) to transfer commands, data, and responses betweenstorage systems104,108. The communication protocol, in one embodiment, is vendor specific and uses SCSI Send/Receive Diagnostics, Read/Write Buffer, and/or vendor specific SCSI operation codes. Of course, it is important to note that even though the protocol can be based upon SCSI, reliance on SCSI as a data transport mechanism is not a requirement of the present invention.
• To perform remote replication in accordance with embodiments of the present invention, the destinationdata storage system108 issues commands to a sourcedata storage system104 requesting the sourcedata storage system104 perform certain tasks (e.g., take snapshots, map snapshots to host LUNs, etc.). The sourcedata storage system104 responds to received commands, performs the requested tasks, and returns the appropriate response. The bridging protocol affords for communications between different types of storage systems. Command sets that may be included as a part of the bridging protocol include, but are not limited to, creating and naming of snapshots, deletion of snapshots, establishment of remote data replication characteristics/parameters, initiation and termination of remote data replication, naming of remote storage volumes, naming of remote snapshot volumes, mapping of remote storage volumes to a LUN, and mapping of remote snapshot volumes to a LUN.
• The create and name a snapshot command is generally used by the destinationdata storage system108 to request the sourcedata storage system104 to take a snapshot of the volume that is targeted during the replication process. The snapshot delete command is used to delete snapshots on the sourcedata storage system104 in order to free-up additional storage resources.
• When the destinationdata storage system108 issues a remote data replication characteristic/parameter command, it is essentially asking the sourcedata storage system104 to provide its replication characteristics, which will ultimately determine how the replication process will proceed between thestorage systems104,108.
• The initiation and termination commands are generally used by the destinationdata storage system108 to specify that replication is starting or terminating at a particular sourcedata storage system104. Alternatively, the initiation and termination commands may be used to specify when the replication process should start/end.
• Storage volume naming commands may be used by the destinationdata storage system108 to name or change the name of a source data storage volume. Likewise, snapshot naming commands may be used by the destinationdata storage system108 to name or change the name of a snapshot volume and may be included as a part of the storage volume naming command.
• The command to map a storage volume to a LUN is generally used by the destinationdata storage system108 to have the sourcedata storage system104 map a storage volume to a LUN. Similarly, the destinationdata storage system108 may use a map snapshot volume to LUN command to have the sourcedata storage system104 map one or more snapshots to a LUN. Also, the map snapshot volume to LUN command may be included in the map storage volume to LUN command.
• The foregoing discussion of the invention has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit the invention to the form disclosed herein. Consequently, variations and modifications commensurate with the above teachings, within the skill and knowledge of the relevant art, are within the scope of the present invention. The embodiments described hereinabove are further intended to explain the best modes presently known of practicing the invention and to enable others skilled in the art to utilize the invention in such, or in other embodiments, and with the various modifications required by their particular application or use of the invention. It is intended that the appended claims be construed to include alternative embodiments to the extent permitted by the prior art.

Claims (20)

1. A method of reproducing data, comprising:

sending a first command from an initiator storage system to a target storage system requesting the target storage system to take a snapshot of data stored on the target storage system;

sending a second command from the initiator storage system to the target storage system requesting the target storage system to make the snapshot viewable to the initiator storage system; and

copying at least a portion of data from the viewable snapshot.

2. The method ofclaim 1, further comprising:

sending a third command from the initiator storage system to the target storage system requesting the target storage system to take a second snapshot of data stored on the target storage system;

sending a fourth command from the initiator storage system to the target storage system requesting the target storage system to make the second snapshot viewable to the initiator storage system; and

copying at least a portion of data from the second viewable snapshot.

3. The method ofclaim 2, further comprising:

identifying at least one difference between the viewable snapshot and the second viewable snapshot;

creating a list that outlines the identified at least one difference; and

updating the copy of the snapshot from the viewable snapshot to the second viewable snapshot by optionally referencing the differences outlined in the list.

4. The method ofclaim 2, wherein the third and fourth commands are sent a predetermined time after the first and second commands are sent.

5. The method ofclaim 1, further comprising the initiator storage system taking a snapshot of the copied data.

6. The method ofclaim 1, further comprising identifying when the snapshot was taken and time stamping the snapshot with the identified time.

7. The method ofclaim 1, wherein at least one of the first and second command comprise a Small Computer Systems Interface (SCSI) command.

8. An electronic data system, comprising:

a target data storage system; and

an initiator data storage system operable to send a first command to the target data storage system requesting the target data storage system to take a snapshot of data stored on the target data storage system, send a second command to the target data storage system requesting the target data storage system to make the snapshot viewable to the initiator data storage system, and copy at least a portion of data from the viewable snapshot.

9. The system ofclaim 8, wherein the initiator data storage system is further operable to send a third command to the target data storage system requesting the target data storage system to take a second snapshot of data stored on the target data storage system, send a fourth command to the target data storage system requesting the target data storage system to make the second snapshot viewable to the initiator data storage system, and copy at least a portion of data from the second viewable snapshot.

10. The system ofclaim 9, wherein the target data storage system is operable to identify at least one difference between the viewable snapshot and the second viewable snapshot, create a list that outlines the identified at least one difference, and present the list to the initiator data storage system, and wherein the initiator data storage system is operable to update the copy of the snapshot from the viewable snapshot to the second viewable snapshot.

11. The system ofclaim 9, wherein the third and fourth commands are sent a predetermined time after the first and second commands are sent.

12. The system ofclaim 8, wherein the target data storage system is further operable to identify when the snapshot was taken and time stamp the snapshot with the identified time.

13. The system ofclaim 8, wherein the second command comprises a command to map the snapshot to a Logical Unit Numbers (LUN) and to make the LUN visible to the initiator data storage system.

14. The system ofclaim 8, wherein the initiator data storage system is further operable to send a command to the target data storage system requesting the target data storage system to determine changes that have occurred since the previous snapshot was taken, send a command to the target data storage system requesting the target data storage system to make the changes visible to the initiator data storage system, and copy the changes that have occurred since the previous snapshot was taken.

15. A device for use in conjunction with data storage, comprising:

an interface for communicating with a target storage system; and

a processor operable to generate a first command for transmission to the target storage system requesting the target storage system to take a snapshot of data stored on the target storage system, generate a second command for transmission to the target data storage system requesting the target data storage system to make the snapshot viewable, and copy at least a portion of data from the viewable snapshot.

16. The device ofclaim 15, wherein the processor is further operable to generate a third command for transmission to the target storage system requesting the target storage system to take a second snapshot of data stored on the target storage system, generate a fourth command for transmission to the target storage system requesting the target storage system to make the second snapshot, and copy at least a portion of data from the second viewable snapshot.

17. The device ofclaim 16, wherein the processor is operable to use a list generated by the target storage system to update the copy of the snapshot, wherein the list outlines at least one difference between the snapshot and the second snapshot.

18. The device ofclaim 16, wherein the third and fourth commands are sent a predetermined time after the first and second commands are sent.

19. The device ofclaim 15, wherein the processor is further operable to take a snapshot of the copied data.

20. The device ofclaim 15, wherein the processor is further operable to generate a command for transmission to the target storage system requesting the target storage system to determine changes that have occurred since the previous snapshot was taken, generate a command for transmission to the target storage system requesting the target storage system to make the changes visible, and update the copied data with the visible changes.

Images