Data vaulting in emergency shutdown

BACKGROUND

p-0003The present invention relates generally to data storage, and particularly to methods and systems for protecting cached data against loss of electrical power.

p-0004Various methods and systems for data storage are known in the art. A typical storage application accepts Input/Output (I/O) commands (e.g., read and write commands) from a host, and executes the commands vis-à-vis a storage device, such as a disk. Some data storage systems employ caching techniques, which cache read and/or write commands in volatile memory before committing them to disk. Caching techniques generally improve the I/O access time of the system and the efficiency of the storage process.

BRIEF SUMMARY

p-0005Embodiments of the present invention provide a method for data storage. The method includes accepting write commands belonging to a storage operation invoked by a host computer, and caching the write commands in a volatile memory that is powered by external electrical power. A current execution status of the storage operation is also cached in the volatile memory.

p-0006Responsively to an interruption of the external electrical power during the storage operation, the cached write commands and the cached execution status are backed-up in a non-volatile memory. Upon resumption of the external electrical power, the backed-up execution status is recovered, so as to resume the interrupted storage operation.

p-0007In some embodiments, the execution status includes an operation journal. Additionally or alternatively, the execution status may include state information of the storage operation. In a disclosed embodiment, the write commands specify data that are sent by the host computer for storage in one or more storage devices, and the storage operation manipulates a logical volume allocated in the one or more storage devices. In some embodiments, the storage operation creates the logical volume, deletes the logical volume, creates a snapshot of the logical volume at a given point in time and/or deletes a previously-created snapshot of the logical volume.

p-0008In another embodiment, the write commands specify data items that are sent by the host computer for storage in one or more storage devices, and backing-up the cached write commands includes backing-up only a subset of the cached write commands, which specify first instances of the data items that are more recent than corresponding second instances of the respective data items that are stored in the storage devices. In yet another embodiment, accepting the write commands includes communicating with the host computer over a Storage Area Network (SAN).

p-0009In still another embodiment, the non-volatile memory includes a storage device powered by a temporary power source. The storage device may include at least one component selected from a group of components consisting of a disk and a Random Access Memory (RAM) device. In some embodiments, the non-volatile memory includes a Flash memory device. In an embodiment, the write commands specify data sent by the host computer for storage in one or more storage devices, and the non-volatile memory includes a memory space that is allocated in the one or more storage devices.

p-0010Apparatus and computer software product for data storage are also described.

p-0011The present invention will be more fully understood from the following detailed description of the embodiments thereof, taken together with the drawings in which:

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

p-0012<figref idrefs="DRAWINGS">FIG. 1</figref> is a block diagram that schematically illustrates a data storage system, in accordance with an embodiment of the present invention;

p-0013<figref idrefs="DRAWINGS">FIG. 2</figref> is a block diagram that schematically illustrates a storage processing module, in accordance with an embodiment of the present invention; and

p-0014<figref idrefs="DRAWINGS">FIG. 3</figref> is a flow chart that schematically illustrates a method for protecting cached data, in accordance with an embodiment of the present invention.

DETAILED DESCRIPTION

Overview

p-0015In some known data storage systems, a storage controller receives write commands from one or more hosts with data for storage in one or more storage devices, e.g., disks. The storage controller initially caches the write commands in a volatile write cache, before the commands are performed on the disks. During normal operation, the storage controller is powered from an external power supply. Since the write cache comprises volatile memory, it is vulnerable to power loss events.

p-0016In order to protect against such events, some known storage configurations use an auxiliary storage device that is powered by a temporary power source, such as a battery. When external power is unavailable, the storage controller immediately backs-up at least some of the contents of the write cache to the auxiliary storage device. During the back-up operation, electrical power to the write cache and auxiliary storage device is provided by the temporary power source. When external power is reapplied, the storage controller recovers the contents of the write cache from the auxiliary storage device. This technique is sometimes referred to as a “fire hose dump.”

p-0017The mere backing-up of write commands, however, is not sufficient for providing seamless recovery of the system following a power loss event. In a typical system, the hosts carry out storage operations, such as “create volume” or “delete volume” operations, which modify the organization, layout or storage configuration of the stored data in the storage devices. When recovering from power loss, such storage operations often have to be canceled, rolled back or restarted, even when the write commands belonging to these operations were fully backed-up and recovered.

p-0018Embodiments of the present invention provide improved methods and systems for protecting against power loss events in data storage applications. In some embodiments that are described hereinbelow, the storage controller caches, in addition to the write commands, the execution status of storage operations that are currently in progress. When backing-up the contents of the write cache to the auxiliary storage device, the storage controller backs-up the current execution status, as well. When external power is reapplied, the storage controller recovers the backed-up information, including the execution status of the different storage operations.

p-0019Thus, when using the methods and systems described herein, the storage controller can resume the storage operations performed by the hosts from the point at which they were disrupted using the recovered execution status, without having to roll-back or restart the operations. As a result, the recovery from power loss events is rapid and incurs only a minimum of additional I/O operations, in comparison with known methods.

System Description

p-0020<figref idrefs="DRAWINGS">FIG. 1</figref> is a block diagram that schematically illustrates a data storage system <b>20</b>, in accordance with an embodiment of the present invention. System <b>20</b> comprises a storage controller <b>24</b>, which stores and retrieves data for hosts <b>28</b>. The hosts are also referred to as initiators.

p-0021In the configuration of <figref idrefs="DRAWINGS">FIG. 1</figref>, the hosts are connected to the storage controller via a Storage Area Network (SAN), as is known in the art. The SAN typically comprises one or more network switches <b>36</b>. The hosts and storage controller may communicate over SAN <b>32</b> using any suitable protocol, such as the Small Computer System Interface (SCSI) and/or Fibre-Channel (FC) protocols. Although the embodiment of <figref idrefs="DRAWINGS">FIG. 1</figref> refers to a SAN configuration, the hosts and storage controller may be connected using any other suitable configuration, such as a Network-Attached Storage (NAS) or Direct-Attached Storage (DAS) configuration.

p-0022Storage controller <b>24</b> comprises multiple storage processing modules <b>40</b>, which store data in multiple storage devices, such as disks <b>44</b>. Storage controller <b>24</b> may comprise any desired number of modules <b>40</b> and any desired number of disks <b>44</b>. In a typical configuration, the storage controller may comprise between 1-32 storage processing modules and between 2-2000 disks, although any other suitable numbers can also be used. In the exemplary configuration of <figref idrefs="DRAWINGS">FIG. 1</figref>, each module <b>40</b> stores data in a separate set of disks <b>44</b>. In alternative embodiments, however, a given disk <b>44</b> need not be uniquely associated with a particular module <b>40</b>. For example, a pool of disks <b>44</b> may be common to all modules <b>40</b>.

p-0023<figref idrefs="DRAWINGS">FIG. 2</figref> is a block diagram that schematically illustrates storage processing module <b>40</b>, in accordance with an embodiment of the present invention. Module <b>40</b> comprises a network interface <b>48</b>, which communicates with SAN <b>32</b>. In particular, the network interface accepts I/O commands (e.g., read and write commands) from hosts <b>28</b> and forwards the commands to other elements of module <b>40</b>.

p-0024Module <b>40</b> comprises a write cache <b>52</b>, which caches some or all of the write commands accepted by the module before storing them to disks <b>44</b>. Each write command comprises one or more memory pages containing data for storage. The terms “storing a write command” or “caching a write command” are used herein to refer to storage or caching of the data contained or indicated in the command, as well as to storage or caching of additional information or instructions conveyed by the command. Such additional information is referred to herein as metadata. While some metadata items may corresponds to a given page, other metadata items may not be page-specific, as will be explained in greater detail below.

p-0025Write cache <b>52</b> comprises a volatile memory, such as static or dynamic Random Access Memory (RAM). Caching of write commands is advantageous for several reasons. For example, the access time of cache <b>52</b> is usually much lower than the access time of disks <b>44</b>. Thus, assuming module <b>40</b> acknowledges each write command as it is cached without waiting for the command to be stored in disks <b>44</b>, the access time seen by hosts <b>28</b> is considerably reduced.

p-0026Moreover, module <b>40</b> may improve the efficiency of the storage process by accumulating multiple commands in cache <b>52</b> and only then storing the accumulated commands in disks <b>44</b>. For example, module <b>40</b> may optimize the order in which the write commands are performed, e.g., in accordance with the target locations in the storage device, in order to reduce storage time. As another example, if a certain data item is addressed by several write commands, module <b>40</b> may carry only the write command corresponding to the most current version of the data item.

p-0027On the other hand, since write cache <b>52</b> comprises volatile memory, it is vulnerable to power loss events. The write cache, and possibly some other elements of module <b>40</b>, operates on external electrical power that is supplied to the module. When this external power supply is unavailable, for any reason, the data cached in write cache <b>52</b> is lost.

p-0028Embodiments of the present invention provide improved methods and systems for protecting the content of the write cache against power interruption. The methods and systems described herein protect the cached information against any scenario in which external power is unavailable. Such scenarios may comprise, for example, power outages, equipment failures or even intentional power shutdown events.

p-0029In order to protect the content of the write cache, module <b>40</b> comprises an auxiliary Storage Device (SD) <b>56</b>, a temporary power source <b>60</b> and a processor <b>64</b>. When external power is unavailable, processor <b>64</b> backs-up some or all of the write commands that are cached in cache <b>52</b> to auxiliary SD <b>56</b>. Temporary power source <b>60</b> provides electrical power to write cache <b>52</b>, auxiliary SD <b>56</b> and processor <b>64</b> for a period of time that is sufficient for backing-up the write commands to the auxiliary SD. When external power is reapplied, processor <b>64</b> reconstructed the content of write cache <b>52</b> from the data backed-up in auxiliary SD <b>56</b>. This process is sometimes referred to as a “fire hose dump” process.

p-0030In some embodiments, each cached page has a corresponding page descriptor, which holds metadata pertaining to the page. For example, the page descriptor may comprise a flag indicating whether the version of the page that is cached in cache <b>52</b> is more recent than the version that is stored in disks <b>44</b>. A page whose cached version is more recent than its stored version is referred to as a “dirty” page. Typically, processor <b>64</b> backs-up only “dirty” pages to the auxiliary SD.

p-0031In a typical configuration, auxiliary SD <b>56</b> comprises a disk and power source <b>60</b> comprises a suitable battery. In some embodiments, the auxiliary SD comprises a disk that is separate from disks <b>44</b>. In alternative embodiments, a certain storage space on disks <b>44</b> is allocated for use as the auxiliary SD. This storage space may comprise any suitable storage space on disks <b>44</b>, e.g., an entire disk, several disks, or one or more fractions of one or more disks. Further alternatively, auxiliary SD <b>56</b> may comprise a non-volatile memory such as a Flash memory. In yet another embodiment, the auxiliary SD may comprise one or more RAM devices that are powered by temporary power source <b>60</b>.

p-0032The energy capacity of power source <b>60</b>, the storage size of auxiliary SD <b>56</b> and the storage size of write cache <b>52</b> are often interrelated. Typically, the auxiliary SD is dimensioned to have sufficient size for holding the “dirty” pages cached in write cache <b>52</b>. Power source <b>60</b> is specified to have sufficient energy for powering the processor, the write cache and the auxiliary SD for the entire duration of the back-up process.

p-0033The configuration of module <b>40</b> shown in <figref idrefs="DRAWINGS">FIG. 2</figref> is an exemplary configuration, which was chosen purely for the sake of conceptual clarity. Elements of module <b>40</b> that are not necessary for explaining the methods and systems described herein were omitted from the figure for clarity. For example, module <b>40</b> often comprises a read cache for caching read commands, which is not shown in the figure. In alternative embodiments, module <b>40</b> may have any other suitable configuration.

Backing-Up the Status of Interrupted Storage Operations

p-0034In a typical application, hosts <b>28</b> carry out storage operations comprising I/O (read and write) commands. In other words, each write command can be viewed as a building block of a higher-level storage operation carried out by a host. The storage operations typically comprise operations that modify the organization, layout or storage configuration of the data that is stored in disks <b>44</b>.

p-0035In many cases, a storage operation addresses or manipulates a specific logical volume allocated in storage controller <b>24</b>. A logical volume, as is well-known in the art, comprises a virtual storage unit assigned in storage controller <b>24</b>. Volumes are also referred to as Logical Units (LU) having Logical Unit Numbers (LUN). A storage operation carried out by a host may comprise, for example, a “create volume,” “delete volume,” “resize volume” or “format volume” operation, each of which can be represented as a sequence of read and/or write operations.

p-0036Another class of storage operations addresses the manipulation of snapshots. A snapshot is a copy of a given volume, which is created at a certain point in time. Storage operations that manipulate snapshots may comprise, for example, “create snapshot” and “delete snapshot” operations. Yet another class of storage operations manipulates remote copies in mirroring applications, i.e., copies of logical volumes that are stored in a remote storage controller. Operations in this class may comprise, for example, “create remote copy” and “delete remote copy” operations. Further alternatively, hosts <b>28</b> may carry out any other suitable storage operation that modifies the layout, organization or configuration of the stored data and can be represented by a sequence of read and write commands.

p-0037Some storage operations may comprise lower-level steps, such as updating hierarchical data structures used for representing logical volumes and snapshots, updating an operation journal that logs the operation progress, updating a bitmap of physical partitions in disks <b>44</b>, transferring of in-progress transactions to a remote mirror, and/or organization of free storage space (“garbage collection”).

p-0038As can be appreciated, when one or more write commands belonging to a certain storage operation are cached in cache <b>52</b>, unavailability of external power in module <b>40</b> may disrupt the entire storage operation. When external power is resumed, the storage operation may need to be canceled, rolled-back or restarted.

p-0039Embodiments of the present invention provide improved methods and systems for protecting against disruption of storage operations caused by loss of external power. In addition to caching data and metadata, the methods and systems described herein cache in write cache <b>52</b> the current execution status of the storage operations that are currently in progress. The current execution status may comprise an operation journal and/or state information regarding storage operations that are in progress. In response to detecting that external power is unavailable, processor <b>64</b> backs-up the specification of the current execution status in auxiliary SD <b>56</b>.

p-0040For a given storage operation, the metadata and execution status backed-up in the auxiliary SD are sufficient for resuming the storage operation from the point it was disrupted. When external power is reapplied, processor <b>64</b> recovers the backed-up metadata and execution status, and uses this information to resume the storage operation from the point it was disrupted by the power loss event. Thus, when using this technique, recovery from power loss events is relatively quick and incurs only a minimum of extra I/O commands.

p-0041The execution status may comprise any suitable parameters regarding the storage operations and/or their execution state. For example, the metadata may record the type of storage operation that is carried out, the parameters provided as input to the operation (e.g., an identifier of a logical volume that is to be modified by the operation and/or a new desired size for the logical volume), the execution stage that the operation has reached (in other words, what metadata has started changing). In some embodiments, the execution status comprises an operation journal, which records which operations were received for execution, the parameters provided as input to the operations, and the last state the operations were in. This sort of information enables the storage controller to seamlessly resume any interrupted operations when external power is reapplied.

Data Vaulting Method Description

p-0042<figref idrefs="DRAWINGS">FIG. 3</figref> is a flow chart that schematically illustrates a method for protecting cached data, in accordance with an embodiment of the present invention. The method begins with module <b>40</b> of storage controller <b>24</b> receiving write commands from one or more hosts <b>28</b>, at a command acceptance step <b>70</b>. Each write command belongs to a certain storage operation that is invoked by a certain host. As explained above, the storage operations modify the layout or organization of the stored data. Thus, step <b>70</b> can be viewed as a step of receiving both commands for writing data and commands that modify the data configuration.

p-0043Processor <b>64</b> of module <b>40</b> caches the write commands, as well as applicable metadata and execution status, in write cache <b>52</b>, at a caching step <b>74</b>. In particular, processor <b>64</b> caches the current execution status of the storage operations that are currently in progress. At a later point in time, processor <b>64</b> transfers the cached write commands for storage in disks <b>44</b>.

p-0044Processor <b>64</b> checks whether external power is available, at a power loss checking step <b>82</b>. In some embodiments, module <b>40</b> receives an alarm signal (e.g., from an external Uninterruptible Power Supply—UPS) indicating external power loss and an approaching shutdown. In the Advanced Interactive executive (AIX) operating system, this signal is denoted Early Power Off Warning (EPOW). Unix-based systems may use a configuration file called psmon.conf for monitoring power supply status. Alternatively, processor <b>64</b> may use any other suitable signal or indication to check the status of the external power supply.

p-0045If external power is available, the method loops back to step <b>70</b> above, and module <b>40</b> continues to receive, cache and store write commands. If processor <b>64</b> detects that external power is unavailable, the processor immediately backs-up the “dirty” pages from write cache <b>52</b> to auxiliary SD <b>56</b>, at a fire hosing step <b>86</b>. In addition to the “dirty” pages, processor <b>64</b> also copies the metadata and the current execution status of the storage operations that are currently in progress. In some embodiments, the backed-up execution status comprises an operation journal and operation-state data, which records the current state of each storage operation. When backing-up the data, metadata and execution status to the auxiliary SD, electrical power to the processor, auxiliary SD and write cache is provided by temporary power source <b>60</b>.

p-0046Processor <b>64</b> checks whether external power is reapplied, at a power status checking step <b>90</b>. Upon detecting that power is reapplied, the processor recovers the data, metadata and execution status that is backed-up in auxiliary SD <b>94</b> to the write cache, at a recovery step <b>94</b>. In particular, the processor recovers the backed-up execution status of the storage operations at the point in time in which external power was disrupted.

p-0047Using the recovered execution status, processor <b>64</b> resumes the execution of each disrupted storage operation from the state in which it was disrupted, at an operation resumption step <b>98</b>.

p-0048As will be appreciated by one skilled in the art, the present invention may be embodied as a system, method or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Furthermore, the present invention may take the form of a computer program product embodied in any tangible medium of expression having computer usable program code embodied in the medium.

p-0049Any combination of one or more computer usable or computer readable medium(s) may be utilized. The computer-usable or computer-readable medium may be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or propagation medium. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CDROM), an optical storage device, a transmission media such as those supporting the Internet or an intranet, or a magnetic storage device. Note that the computer-usable or computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory. In the context of this document, a computer-usable or computer-readable medium may be any medium that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-usable medium may include a propagated data signal with the computer-usable program code embodied therewith, either in baseband or as part of a carrier wave. The computer usable program code may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc.

p-0050Computer program code for carrying out operations of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).

p-0051The present invention is described herein with reference to flow chart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flow chart illustrations and/or block diagrams, and combinations of blocks in the flow chart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.

p-0052These computer program instructions may also be stored in a computer-readable medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable medium produce an article of manufacture including instruction means which implement the function/act specified in the flow charts and/or block diagram block or blocks.

p-0053The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flow charts and/or block diagram block or blocks.

p-0054The flow charts and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flow charts or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flow chart illustrations, and combinations of blocks in the block diagrams and/or flow chart illustrations, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

p-0055Although the embodiments described above mainly address back-up and recovery in storage controllers, the methods and systems described herein can also be used in other systems that cache write commands, such as in-memory database systems with Hard Disk Drives (HDD) having battery backup.

p-0056It will thus be appreciated that the embodiments described above are cited by way of example, and that the present invention is not limited to what has been particularly shown and described hereinabove. Rather, the scope of the present invention includes both combinations and sub-combinations of the various features described hereinabove, as well as variations and modifications thereof which would occur to persons skilled in the art upon reading the foregoing description and which are not disclosed in the prior art.