SYNCHRONIZED DATA DUPLICATION
A system and method for data deduplication is presented. Data received from one or more computing systems is deduplicated, and the results of the deduplication process stored in a reference table. A representative subset of the reference table is shared among a plurality of systems that utilize the data deduplication repository. This representative subset of the reference table can be used by the computing systems to deduplicate data locally before it is sent to the repository for storage. Likewise, it can be used to allow deduplicated data to be returned from the repository to the computing systems. In some cases, the representative subset can be a proper subset wherein a portion of the referenced table is identified shared among the computing systems to reduce bandwidth requirements for reference-table synchronization.
Any and all priority claims identified in the Application Data Sheet, or any correction thereto, are hereby incorporated by reference under 37 CFR 1.57.BACKGROUND
1. Technical Field
The present invention generally relates to data deduplication, and more particularly, some embodiments relate to systems and methods for facilitating shared deduplication information.
2. Description of the Related Art
The storage and retrieval of data is an age-old art that has evolved as methods for processing and using data have evolved. In the early 18th century, Basile Bouchon is purported to have used a perforated paper loop to store patterns used for printing cloth. In the mechanical arts, similar technology in the form of punch cards and punch tape were also used in the 18th century in textile mills to control mechanized looms. Two centuries later, early computers also used punch cards and paper punch tape to store data and to input programs.
However, punch cards were not the only storage mechanism available in the mid-20th century. Drum memory was widely used in the 1950s and 1960s with capacities approaching about 10 kb, and the first hard drive was developed in the 1950s and is reported to have used 50 24-inch discs to achieve a total capacity of almost 5 MB. However, these were large and costly systems and although punch cards were inconvenient, their lower cost contributed to their longevity as a viable alternative.
In 1980 the hard drive broke the 1 GB capacity mark with the introduction of the IBM 3380, which could store more than two gigabytes of data. The IBM 3380, however, was about as large as a refrigerator, weighed ¼ ton, and cost in the range of approximately $97,000 to $142,000, depending on the features selected. This is in stark contrast to contemporary storage systems that provide for storage of hundreds of terabytes of data or more for seemingly instantaneous access by networked devices. Even handheld electronic devices such as digital cameras, MP3 players and others are capable of storing gigabytes of data, and today's desktop computers boast hundreds of gigabytes of storage capacity.
However, with the advent of networked computing, storage of electronic data has migrated from the individual computer to network-accessible storage devices. These include, for example, optical libraries, Redundant Arrays of Inexpensive Disks (RAID), CD-ROM jukeboxes, drive pools and other mass storage technologies. These storage devices are accessible to and can be shared by individual computers such as via a Local Area Network (LAN), a Wide Area Network (WAN), or a Storage Area Network (SAN) to name a few. These client computers not only access their own local storage devices but also storage devices of the network to perform backups, transaction processing, file sharing, and other storage-related operations.
The large volumes of data often stored and shared by networked devices can cause overloading of the limited network bandwidth. For example, during operations such as system backups, transaction processing, file copying and transfer, and other similar operations, the communication bandwidth of the network often becomes the rate-determining factor.
In addition, even with large capacity storage systems, computing enterprises are being overloaded by vast amounts of data. Documents sent via email, for example, can be copied and resent multiple times and several instances of the very same document might be stored many times in many different locations. IT administrators are struggling to keep up with the seemingly exponential increase in the volume of documents, media and other data. This problem is severely compounded by other factors such as the large file sizes often associated with multi-media files, and file proliferation through email and other content sharing mechanisms. However, additional storage capacity requires capital expenditures, consumes power, takes up floor space and burdens administrative overhead. Even with additional storage capacity, the sheer volume of data becomes a strain on backup and data recovery plans, leading to greater risk in data integrity.
As an alternative to simply increasing the amount of storage capacity, contemporary enterprises have turned to compression and other like technologies to reduce the volume of data. One such technology that can be used is known as data deduplication. Data deduplication in its various forms eliminates or reduces the amount of redundant data by implementing policies that strive to reduce the quantity of, or even eliminate, instances of redundant data blocks in storage. With data deduplication, data is broken up into segments or blocks. As new data enters the system, the segments are checked to see if they already exist in storage. If a segment already exists, rather than store that segment again, a pointer to the location of the existing segment is stored.
The segment size selected for data deduplication can be defined at various levels, from small segment sizes (for example, 1 kB or less) to much larger segment sizes, and to entire files. A larger segment size can yield greater space or bandwidth savings on a per-instance basis, however, the opportunities for identifying redundancies may be reduced with larger segment sizes. These tradeoffs can depend on the system with which deduplication is implemented and the types of data or files it handles.
As indicated above, in some instances, deduplication can be performed on a file-by-file basis. With such a system, rather than storing multiple copies of the same file, one instance of the file is stored, for example, in a central repository, and pointers to the file are stored in place of the redundant copies. However, deduplication at the file level can suffer in efficiencies as compared to deduplication using smaller segment sizes because even a small change in the file generally requires that an entire copy of the file be re-stored.
In addition to reducing the amount of storage space consumed, data deduplication can also help to relieve congestion on crowded communication pathways. In addition, the more efficient use of disk space can often allow data retention periods to increase, adding more integrity to the enterprise. Data deduplication is frequently used in conjunction with other forms of data reduction, including conventional data compression algorithms and delta difference storage.
Data deduplication often relies on hashing algorithms that hash the data segments to generate an identifying signature for the segments. Accordingly, each segment is processed using the hashing algorithm to generate a hash value. The resultant hash value is compared against hash values stored in a hash table to determine whether the segment already exists. If so, the segment is replaced with a pointer to the entry in the table containing the appropriate hash value or pointing to the location of the data in storage. Otherwise, the new data is stored and its hash value is added to the table along with an address for the data.
Because hash functions are not perfect, the same hash value can in some cases be returned for segments containing different data. When such a false-positive occurs, the system can mistake new data for already-stored data and fail to store the new segment. Accordingly, multiple hash algorithms and other techniques can be employed to reduce the likelihood of these so-called hash collisions.BRIEF SUMMARY OF EMBODIMENTS OF THE INVENTION
According to various embodiments, systems and methods are provided for data deduplication. Particularly, in some embodiments, techniques for performing reference table distribution and synchronization are provided. Accordingly, a reference table generated as a result of the deduplication process at a storage repository can be shared among a plurality of client systems that utilize a repository for data storage. This can be implemented to allow the client systems to perform local data deduplication before their data is sent to the repository. Likewise, this can also allow the client systems to receive deduplicated data from the storage repository. Accordingly, systems and methods can be implemented to allow deduplicated data to be transferred among a plurality of computing systems thereby reducing bandwidth requirements for data storage and retrieval operations.
In some embodiments, rather than distribute the entire reference table to each client for synchronization, a proper subset of reference table entries can be identified and shared with the client devices for synchronization. This can be implemented so as to reduce the amount of bandwidth required to synchronize the reference table among the computing systems. In further embodiments, the subset can be identified based on data utilization criteria.
According to an embodiment of the invention, systems and methods are provided for performing data deduplication for data used by a plurality of computing systems. The systems and methods can be configured to perform the steps of receiving at a shared storage repository data from the plurality of computing systems, performing a data deduplication operation on the received data, and transmitting an instantiation of a reference table for the deduplication to determined ones of the plurality of computing systems to allow deduplication to be performed by the determined ones of the plurality of computing systems.
The deduplication operation can include defining a segment of the received data; applying an algorithm to the defined data segment to generate a signature for the defined data segment; comparing the signature for the defined data segment with one or more signatures stored in a reference table for one or more previously defined data segments to determine whether the defined segment is already stored in the shared storage repository; and updating the reference table to include the signature for the defined data segment and a reference for the defined data segment if the defined data segment is not in the shared storage repository.
In one embodiment, a first instantiation of the reference table is sent to a first group of one or more of the plurality of computing systems and a second instantiation of the reference table is sent to a second group of one or more of the plurality of computing systems, wherein the first instantiation of the reference table is different from the second instantiation of the reference table.
The operation can further include a step of determining the instantiation of the reference table to be transmitted, and wherein the instantiation of the reference table is a proper subset of the reference table. The step of determining the instantiation of the reference table can include selecting one or more entries of the reference table based on at least one of utilization rate of data segments represented by the entries and size of the data segments represented by the entries. In another embodiment, the step of determining the instantiation of the reference table comprises selecting one or more entries of the reference table based on a combination of utilization rate of data segments represented by the entries and size of the data segments represented by the entries. Any of a number of combinations can be used. For example, the combination can be a weighted combination of utilization rate of data segments represented by the entries and size of the data segments represented by the entries.
Other features and aspects of the invention will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, which illustrate, by way of example, the features in accordance with embodiments of the invention. The summary is not intended to limit the scope of the invention, which is defined solely by the claims attached hereto.
The present invention, in accordance with one or more various embodiments, is described in detail with reference to the following figures. The drawings are provided for purposes of illustration only and merely depict typical or example embodiments of the invention. These drawings are provided to facilitate the reader's understanding of the invention and shall not be considered limiting of the breadth, scope, or applicability of the invention. It should be noted that for clarity and ease of illustration these drawings are not necessarily made to scale.
The figures are not intended to be exhaustive or to limit the invention to the precise form disclosed. It should be understood that the invention can be practiced with modification and alteration, and that the invention be limited only by the claims and the equivalents thereof.DETAILED DESCRIPTION OF THE EMBODIMENTS OF THE INVENTION
The present invention is directed toward a system and method for data deduplication, and more particularly various embodiments are directed toward systems and methods for synchronization of reference tables to facilitate data deduplication. In various embodiments, data is stored for a plurality of clients in a shared storage environment, and rather than transfer large amounts of data among the clients and the shared storage, the data in shared storage is deduplicated and the hash table or other reference table for the data is shared among some or all of the clients. When a client has data to transfer to or place in the shared storage, that client can run a deduplication algorithm on segments of the data and use its own representative instantiation of the reference table to determine whether the data segments already exist in a shared data store. Accordingly, for a given segment, the client can determine whether to send the entire data segment to the shared storage or just send a reference or pointer or other information from the reference table if the segment is duplicative of what is already in the data store. In a situation where the analyzed segment is not in the data store, the client device can send the hash value or other reference table information to the central storage (or other location maintaining the main reference table) so that the primary reference table can be updated with the information on the newly added segment.
In various embodiments, the data store or shared storage repository can comprise any of a number of data storage architectures. For example, in one application, the shared storage can comprise one or more data storage systems accessible by and shared among multiple client systems such as, for example, one or more dedicated storage repositories or centralized storage repositories. In another example, the shared storage repository can comprise a plurality of storage locations distributed across some or all of the multiple clients among which the data is shared, or a combination of distributed and centralized storage devices.
In further embodiments of the invention, rather than send an entire reference table to each of the clients or client groups that are sharing the common storage, a subset of the table can be identified and only that subset is transmitted to the client systems to synchronize the reference tables among all of the devices. Additionally, different client systems or groups of client systems can receive different subsets of the reference table. Subsets can be defined for a given client or group of clients based on data utilization or other factors. Transferring a subset of the reference table rather than the entire reference table can reduce bandwidth consumption across the network, result in increased efficiency and reduce synchronization problems with the database.
For clarification, in the various embodiments described herein, the term synchronization is not intended to require that all client devices be updated with a representative instantiation of the reference table at the same time or that all client devices receive the same reference table. Although in some embodiments the reference table or updates thereto can be broadcast to all participating systems simultaneously (or close to simultaneously), in other embodiments the reference table or its updates can be sent to different client devices or groups of client devices at different times. Likewise, in some embodiments the same reference table, or the same subset can be sent to all clients. However, in other embodiments, subsets can be tailored for a given client or group of clients.
Before describing the invention in detail, it is useful to describe a few example environments with which the invention can be implemented. One such example is that of a straightforward data deduplication algorithm with which the systems and methods described herein can be implemented.
Referring now to
On the other hand, if the resultant hash value for a given segment already exists in the hash table 127, this indicates that the same data sequence in that segment is already stored in file system 124. Accordingly, rather than storing the entire segment in the file, only the pointer or address to the same segment that was previously stored needs to be retained. Following this methodology for data stream 120 results in the example file 129 illustrated as being stored in file system 124. This example file includes the first instances of segments A, B, C and D. However, for the subsequent instances of segments A and C that occurred in the incoming data stream 120, the file includes a pointer to the originally stored segments A and C. This is illustrated by ADDRESS A and ADDRESS C in file system 124.
To re-create the original data stream, segments are retrieved from file system 124 and assembled in order. Where an address exists in place of the actual data elements of the segment (ADDRESS A and ADDRESS C in the instant example), that address is accessed, the data retrieved and packaged into the reassembled data stream. In this example, resultant data stream 121 contains the same data as existed in original data stream 120.
Although the illustrated example depicts a system that utilizes a simple reference table having a hash value and pointer value for each segment, more complex systems can also make up an environment for the systems and methods described herein. For example, for the hash values the reference table can also include the source or sources of the data segment represented by the hash, a counter of the number of times that a given segment is encountered, the location of where the segments occur on client devices, and so on. As such, the reference table can be implemented as a hybrid of a hash table and a file allocation table (FAT). This can be useful as a backup in the event that a client system crashes or otherwise goes off line.
The above example is described in terms of an individual computing system (having one or more computing devices) performing local data deduplication for local data storage using a hash function. Data deduplication can also be performed for a plurality of computing systems using shared or local data storage or a combination thereof. For example, the data segments need not be stored at a central location such as file system 124 but can be stored at one or more client locations or at a combination of client locations and central storage locations. Accordingly, the pointers or addresses stored in the reference table can point to the actual storage location of the referenced segment whether that location be at a client storage location or in a central storage repository.
In addition, techniques other than hash functions can be used for data deduplication. Other algorithms can be used to generate a signature for the blocks of data. Likewise, other deduplication methods can also be used to identify redundancies or duplicate entries. Accordingly, the terms hash table, signature table, or reference table might be used in this document interchangeably to refer to the table, index, or other like mechanism used to track the data deduplication process, regardless of the actual file structure and regardless of the function used to arrive at the signatures.
As illustrated in the example of
In this example, storage area network 128 can include a plurality of data storage devices 122 to provide sufficient quantities of data storage for the networked computing systems 132. For example, hard disk drives, disk arrays, optical storage drives and other high-volume memory or storage devices can be included with storage area network 128 to provide desired amounts of data storage at specified access speeds. Similarly, network attached storage can include any variety of data storage devices 122 to provide sufficient quantities of data storage at desired access speeds. Illustrated in this example, network attached storage 137 includes removable storage media 124, although fixed media can also be used. Likewise, data storage 122 associated with storage area network 128 can also use fixed or removable media.
Computing systems 132 connected to networks 147, 149 typically include application software 122 to perform desired operations. Although not illustrated, computing systems 132 typically also include an operating system on which the application software 122 runs. The file system 124 can be provided to facilitate and control file access by the operating system and application software 122. File systems 122 can facilitate access to local and remote storage devices for file or data access and storage. As also illustrated, computer systems 132 can include local storage such as a media module media drive 126 with fixed or removable media 136.
In one example, as data is received from a computer system 132 for storage, repository system 151 performs the data deduplication in an in-line or post-processing methodology for storage. For example, in terms of the exemplary deduplication methodology described above with respect to
If, on the other hand, the signature value does not already exist in table 127, the bit sequence is not in data storage. In such a case, this segment is placed into storage and the signature is placed in a new entry in table 127 along with a pointer to the storage location of the new segment. The reference table 127 can also include additional information reflecting information about the data segment such as, for example, the source of this first instance of the data segment, the location of where the segment existed on the client, and a reference counter indicating the number of times the segment was encountered.
As the example of
Sharing the reference table with the client computing systems 132 allows the client systems 132 to deduplicate the data before it is passed to repository system 151 for storage. Further, this can allow the client systems 132 to consider segments already stored in the repository system 151 when doing their local deduplication. With such an arrangement, and assuming again the example duplication process described above with respect to
For large networks or other large computing environments, the hash table or other reference table 127 can grow to be quite large. Where this is the case, a substantial amount of bandwidth can be consumed by synchronizing the reference table 127 amongst repository system 151 and the various client computing systems 132. This situation can be compounded where large amounts of data are being stored by a large number of computing systems 132. In addition, as the repository is pruned, further reference-table synchronization opportunities are presented, leading to additional bandwidth consumption. Accordingly, embodiments can be implemented wherein a proper subset or portion of the reference table 127 is shared among the multiple computing systems to reduce, minimize or avoid sending the entire reference table 127 to the various systems for synchronization. For example, in one implementation, an entire reference table 127 can be sent to the client devices initially, and updates to synchronize the table to account for ongoing changes can be done by sending a proper subset containing less than the entire original table. As another example, in another embodiment, a proper subset is defined and sent initially rather than sending the entire table. Then, updates to the subset are made on an ongoing basis.
Referring now to
As illustrated at step 185, as additional data is received by storage repository 151 and other data removed from storage repository 151, the reference table is updated and maintained by storage repository 151. For example, new entries to the reference table are made to include signatures and pointers for new data segments received and old signatures and pointers are removed from the table as segments are deleted from the repository.
At step 188, the system performs reference table selection to identify a subset of entries in the reference table for streamlined reference-table-synchronization operations. Various algorithms or processes can be used to identify or define a subset of entries for streamlined synchronization. Generally, in one embodiment, algorithms are implemented to identify those data segments being stored in repository system 151 that have the highest utilization for highest likelihood of occurrence. In such an embodiment, synchronization of a relatively small portion of the reference table can result in bandwidth savings of a relatively larger proportion.
At step 192, the reference table is synchronized with one or more client devices 132. In the illustrated example, it is the subset identified in step 188 that is shared with (for example, sent to) client devices 132 to synchronize or update their local instances of the reference table. The process can continue through subsequent deduplication operations in which the reference table is updated with new entries, relevant subsets of the entries are identified in light of the changes to the reference table, and synchronization performed based on the reference table subset. This is indicated by step 195, which shows the reference-table subset generation and synchronization being performed on a periodic basis. In addition to adding new entries to the reference table for newly stored segments, updates to the reference table can also include updates to reference counter values, sources of data segments, and so on. In one embodiment, the synchronization operations can be run on a periodic basis based on temporal criteria such as the expiration of a period of time, or they can be triggered based on throughput metrics or other criteria. Examples of throughput criteria can include criteria such as, for example, the amount of new data stored in or removed from the system, the number of updates made to the reference table, and so on.
As indicated above with reference to step 188, one criteria that can be used to define the relevant subsets for reference table is based on data utilization. For example, in one embodiment, the system tracks not only the existence of a segment in the repository but also the utilization of each of the segments. One way in which utilization can be tracked is by tracking the quantity or frequency of occurrences of a given segment or the number of times it is accessed. The segments can be scored based on the utilization or access rates and ranked accordingly. This can be used in one embodiment to define or identify segments whose signatures will appear on the reference table subset.
As one example, the number of times that a given segment is presented to repository 151 for storage can be tracked by the system. This number can, in many applications, be directly proportional to the amount of communication bandwidth that is being consumed by transferring the segment from client devices 132 to storage repository 151. Accordingly, these higher utilization segments tend to have a larger impact on system bandwidth than segments within lower utilization. In such environments, defining the reference table subset based on utilization can allow a trade-off between to be made reference table size and marginal improvements in bandwidth savings. In other words, where a reference table for synchronization includes entries for infrequently used data segments, inclusion of these entries in the table for synchronization could consume more bandwidth than is saved by allowing these entries to be used for deduplication at the client side.
Another way to define a subset of the reference table for table synchronization can be to identify changes to the table since the last synchronization operation, such as new table entries or deletions. With changes identified, the system can be configured to send only those changes to the clients to update their local copies of the table. As noted above, in some embodiments not all clients are updated at the same time. Accordingly, changes to the reference table can be tracked on a client-by-client basis, or on the basis of groups of clients so that the updates can be managed based on the actual need of given clients.
In various environments, the process might be further streamlined by defining the reference table subset differently for different clients or different groups of clients 132. Thus, for example, different subsets of reference table 211 can be sent to different computing devices resulting in some or all of the client devices 132 having a reference table that is different from other client devices 132. As another example, client devices 132 can be combined into groupings based on various characteristics (described below) and different subsets of reference table 211 can be sent to the different groupings of client devices.
A number of factors or criteria can be considered when identifying targeted reference tables for a client and a group of clients. For example, the system can be configured to analyze the traffic received from each client or client grouping and perform client-specific utilization measurements. The most utilized entries can be identified on a client-by-client basis and the reference table subset identified accordingly. In another example, the reference table subsets for particular clients or groups of clients can be identified based on a number of other criteria including, for example, they size of the segments utilized by each of the clients, the type of processes being performed by each client, the client environment, characteristics of the client system, and so on. Each of these may have an effect on the quantity and type of data to be stored. As noted above, in some embodiments data in the reference table can be included to indicate the source of the data, its storage location, the number of occurrences and so on. Such data can be used in making the determination as to which clients or groups of clients will receive which subsets of the reference table.
As stated above, both segment size and utilization can be used as a metric to identify a subset of entries for the targeted reference tables for a client or group of clients. For example, in embodiments where stored segments can be of different sizes, selecting reference table data for larger segments would generally result in a greater gain in efficiencies than sharing reference data for smaller segments. Likewise, selecting reference table data for more frequently encountered segments generally results in a greater gain in efficiencies than sharing of reference data for infrequently used data. However, these generalizations are not always without exception. Consider for example a large segment of data that is very rarely used, or consider the opposite case of a smaller segment of data that is frequently utilized by a client or group of clients. In the first case, selecting the rarely used large segment for the subset may not yield the greatest gain in efficiencies. Therefore, a combination of factors, such as a combination of object size and utilization can be used to determine the subset of entries for synchronization.
Such a combination can be made, for example, by multiplying the size of an object by its utilization frequency. As another example such a combination can be made as a weighted combination of utilization frequency and object size. As a further illustration of this latter example, objects can be weighted by their size, with larger objects being weighted higher than smaller objects. Their weight can then be multiplied by their utilization frequency to rank them for subset selection. Weighting can be done on a sliding scale or it can be done in groupings based on data object sizes in the system. As an example of such a grouping, the top 20% of objects by size can be given a weighting of 100%, the next 20% by size a weighting of 80% and so on. As another example, size thresholds can be defined such that objects above a certain size threshold can be given a certain weighting. As a specific example of this, objects above 1 MB might be weighted 100%, objects between 750 kB and 1 MB weighted 90% and so on. As these examples illustrate, there are a number of specific implementations that can be used to define a subset based on a combination of factors such as block size and frequency of utilization.
At step 240, the reference table is monitored as data is received and deduplicated. The table is checked to determine whether changes are made to the reference table as a result of the new data. This check can also be made for data that is deleted from storage. This monitoring can be performed at the client side, the central repository, or at both locations. If changes are made, the process of identifying the subset and resynchronizing one or more of the clients with the new subset can be repeated as illustrated by step 244. In one example, the repository can receive data from a client, process and deduplicate the data for storage, and identify and send a representative hash table to the client for use in future storage operations. Because the data and hence the reference table can change over time, the subset can be re-defined and redistributed to the clients. As noted above, in one embodiment the subset is resent while in another embodiment only deltas to the subset are sent to the client to update the client's reference table.
The deduplication table subsets defined in the various embodiments can be used by the client devices for deduplication of data before it is sent to a central repository for storage. Because reference table updates can be client specific and because they can also be sent to different clients at different times, in some embodiments the synchronization does not result in an exact copy of the reference table being instantiated at each client device. Instead, at any given time, different clients can have different instantiations of the reference table and these instantiations are preferably selected based on size, utilization and other factors specific to the client or client group.
As used herein, the term module might describe a given unit of functionality that can be performed in accordance with one or more embodiments of the present invention. As used herein, a module might be implemented utilizing any form of hardware, software, or a combination thereof. For example, one or more processors, controllers, ASICs, PLAs, logical components, software routines or other mechanisms might be implemented to make up a module. In implementation, the various modules described herein might be implemented as discrete modules or the functions and features described can be shared in part or in total among one or more modules. In other words, as would be apparent to one of ordinary skill in the art after reading this description, the various features and functionality described herein may be implemented in any given application and can be implemented in one or more separate or shared modules in various combinations and permutations. Even though various features or elements of functionality may be individually described or claimed as separate modules, one of ordinary skill in the art will understand that these features and functionality can be shared among one or more common software and hardware elements, and such description shall not require or imply that separate hardware or software components are used to implement such features or functionality.
Where components or modules of the invention are implemented in whole or in part using software, in one embodiment, these software elements can be implemented to operate with a computing or processing module capable of carrying out the functionality described with respect thereto. One such example-computing module is shown in
Referring now to
Computing module 300 might include, for example, one or more processors, controllers, control modules, or other processing devices, such as a processor 304. Processor 304 might be implemented using a general-purpose or special-purpose processing engine such as, for example, a microprocessor, controller, or other control logic. In the example illustrated in
Computing module 300 might also include one or more memory modules, simply referred to herein as main memory 308. For example, preferably random access memory (RAM) or other dynamic memory might be used for storing information and instructions to be executed by processor 304. Main memory 308 might also be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 304. Computing module 300 might likewise include a read only memory (“ROM”) or other static storage device coupled to bus 302 for storing static information and instructions for processor 304.
The computing module 300 might also include one or more various forms of information storage mechanism 310, which might include, for example, a media drive 312 and a storage unit interface 320. The media drive 312 might include a drive or other mechanism to support fixed or removable storage media 314. For example, a hard disk drive, a floppy disk drive, a magnetic tape drive, an optical disk drive, a CD or DVD drive (R or RW), or other removable or fixed media drive might be provided. Accordingly, storage media 314, might include, for example, a hard disk, a floppy disk, magnetic tape, cartridge, optical disk, a CD or DVD, or other fixed or removable medium that is read by, written to or accessed by media drive 312. As these examples illustrate, the storage media 314 can include a computer usable storage medium having stored therein computer software or data.
In alternative embodiments, information storage mechanism 310 might include other similar instrumentalities for allowing computer programs or other instructions or data to be loaded into computing module 300. Such instrumentalities might include, for example, a fixed or removable storage unit 322 and an interface 320. Examples of such storage units 322 and interfaces 320 can include a program cartridge and cartridge interface, a removable memory (for example, a flash memory or other removable memory module) and memory slot, a PCMCIA slot and card, and other fixed or removable storage units 322 and interfaces 320 that allow software and data to be transferred from the storage unit 322 to computing module 300.
Computing module 300 might also include a communications interface 324. Communications interface 324 might be used to allow software and data to be transferred between computing module 300 and external devices. Examples of communications interface 324 might include a modem or softmodem, a network interface (such as an Ethernet, network interface card, WiMedia, IEEE 802.XX or other interface), a communications port (such as for example, a USB port, IR port, RS232 port Bluetooth® interface, or other port), or other communications interface. Software and data transferred via communications interface 324 might typically be carried on signals, which can be electronic, electromagnetic (which includes optical) or other signals capable of being exchanged by a given communications interface 324. These signals might be provided to communications interface 324 via a channel 328. This channel 328 might carry signals and might be implemented using a wired or wireless communication medium. These signals can deliver the software and data from memory or other storage medium in one computing system to memory or other storage medium in computing system 300. Some examples of a channel might include a phone line, a cellular link, an RF link, an optical link, a network interface, a local or wide area network, and other wired or wireless communications channels.
In this document, the terms “computer program medium” and “computer usable medium” are used to generally refer to physical storage media such as, for example, memory 308, storage unit 320, and media 314. These and other various forms of computer program media or computer usable media may be involved in storing one or more sequences of one or more instructions to a processing device for execution. Such instructions embodied on the medium, are generally referred to as “computer program code” or a “computer program product” (which may be grouped in the form of computer programs or other groupings). When executed, such instructions might enable the computing module 300 to perform features or functions of the present invention as discussed herein.
While various embodiments of the present invention have been described above, it should be understood that they have been presented by way of example only, and not of limitation. Likewise, the various diagrams may depict an example architectural or other configuration for the invention, which is done to aid in understanding the features and functionality that can be included in the invention. The invention is not restricted to the illustrated example architectures or configurations, but the desired features can be implemented using a variety of alternative architectures and configurations. Indeed, it will be apparent to one of skill in the art how alternative functional, logical or physical partitioning and configurations can be implemented to implement the desired features of the present invention. Also, a multitude of different constituent module names other than those depicted herein can be applied to the various partitions. Additionally, with regard to flow diagrams, operational descriptions and method claims, the order in which the steps are presented herein shall not mandate that various embodiments be implemented to perform the recited functionality in the same order unless the context dictates otherwise.
Although the invention is described above in terms of various exemplary embodiments and implementations, it should be understood that the various features, aspects and functionality described in one or more of the individual embodiments are not limited in their applicability to the particular embodiment with which they are described, but instead can be applied, alone or in various combinations, to one or more of the other embodiments of the invention, whether or not such embodiments are described and whether or not such features are presented as being a part of a described embodiment. Thus, the breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments.
Terms and phrases used in this document, and variations thereof, unless otherwise expressly stated, should be construed as open ended as opposed to limiting. As examples of the foregoing: the term “including” should be read as meaning “including, without limitation” or the like; the term “example” is used to provide exemplary instances of the item in discussion, not an exhaustive or limiting list thereof; the terms “a” or “an” should be read as meaning “at least one,” “one or more” or the like; and adjectives such as “conventional,” “traditional,” “normal,” “standard,” “known” and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass conventional, traditional, normal, or standard technologies that may be available or known now or at any time in the future. Likewise, where this document refers to technologies that would be apparent or known to one of ordinary skill in the art, such technologies encompass those apparent or known to the skilled artisan now or at any time in the future.
The presence of broadening words and phrases such as “one or more,” “at least,” “but not limited to” or other like phrases in some instances shall not be read to mean that the narrower case is intended or required in instances where such broadening phrases may be absent. The use of the term “module” does not imply that the components or functionality described or claimed as part of the module are all configured in a common package. Indeed, any or all of the various components of a module, whether control logic or other components, can be combined in a single package or separately maintained and can further be distributed in multiple groupings or packages or across multiple locations.
Additionally, the various embodiments set forth herein are described in terms of exemplary block diagrams, flow charts and other illustrations. As will become apparent to one of ordinary skill in the art after reading this document, the illustrated embodiments and their various alternatives can be implemented without confinement to the illustrated examples. For example, block diagrams and their accompanying description should not be construed as mandating a particular architecture or configuration.
2. A computer-implemented data deduplication method, the method comprising:
- with one or more computing systems of a shared storage system comprising computer hardware and in networked communication with a plurality of computing systems that are physically separate from the shared storage system: determining whether a first data segment included in data generated by an application executing on a first computing system of the plurality of computing systems is already stored in the shared storage system; if the first data segment is not already stored in the shared storage system, updating a reference table to include an entry corresponding to the first data segment; based on an analysis of data traffic received from one or more of the plurality of computing systems, determining at least a second computing system of the plurality of computing systems to which to transmit an updated partial instantiation of the reference table that includes the entry corresponding to the first data segment; and transmitting the updated partial instantiation of the reference table from the shared storage system to the second computing system of the plurality of computing systems such that, subsequent to said transmitting, a partial instantiation of the reference table local to the second computing system includes the entry corresponding to the first data segment and a partial instantiation of the reference table local to a third computing system of the plurality of computing systems does not include the entry corresponding to the first data segment.
3. The method of claim 2, wherein the partial instantiations of the reference table local to the second and third computing systems are proper subsets of the reference table.
4. The method of claim 2, further comprising determining additional entries to include in the updated partial instantiation of the reference table based on at least one of data segment utilization rate and data segment size.
5. The method of claim 2, further comprising determining additional entries to include in the partial instantiation of the reference table based on a combination of data segment utilization rate and data segment size.
6. The method of claim 5, wherein the combination comprises a weighted combination of data segment utilization rate and data segment size.
7. The method of claim 2 wherein said determining is in response to receiving the first data segment from the first computing system at the shared storage system.
8. The method of claim 7 further comprising storing the first data segment in the shared storage system.
9. The method of claim 2 further comprising, subsequent to said transmitting, receiving a signature corresponding to the first data segment from the second computing system at the shared storage system, without receiving the first data segment itself.
10. A system, comprising:
- a shared storage repository comprising computer memory; and
- a server system including one or more computing devices comprising computer hardware, the server system in networked communication with a plurality of computing systems which are physically separate from the server system, the server system configured to: determine whether a first data segment included in data generated by an application executing on a first computing system of the plurality of computing systems is already stored in the shared storage repository; if the first data segment is not already stored in the shared storage repository, update a reference table to include an entry corresponding to the first data segment; based on an analysis of data traffic received from one or more of the plurality of computing systems, determine at least a second computing system of the plurality of computing systems to which to transmit an updated partial instantiation of the reference table that includes the entry corresponding to the first data segment; and transmit the updated partial instantiation of the reference table from the server system to the second computing system of the plurality of computing systems such that, subsequent to the transmission of the updated partial instantiation of the reference table, a partial instantiation of the reference table local to the second computing system includes the entry corresponding to the first data segment and a partial instantiation of the reference table local to a third computing system of the plurality of computing systems does not include the entry corresponding to the first data segment.
11. The system of claim 10, wherein the partial instantiations of the reference table local to the second and third computing systems are proper subsets of the reference table.
12. The system of claim 10, wherein the server system is further configured to determine additional entries to include in the updated partial instantiation of the reference table based on at least one of data segment utilization rate and data segment size.
13. The system of claim 10, wherein the server system is further configured to determine additional entries to include in the partial instantiation of the reference table based on a combination of data segment utilization rate and data segment size.
14. The system of claim 13, wherein the combination comprises a weighted combination of data segment utilization rate and data segment size.
15. The system of claim 10 wherein the server system receives the first data segment from the first computing system prior to determining whether the first data segment is included in the data generated by the application.
16. The system of claim 15 wherein the server system is further configured to store the first data segment in the shared storage repository subsequent to receipt from the first computing system.
17. The system of claim 10 wherein subsequent to transmission of the partial instantiation of the reference table to the second computing system, the server system receives from the second computing system a signature corresponding to the first data segment, and wherein the server system is configured to update the reference table in response to receipt of the signature.