US 8,364,652 B2Grant

Content aligned block-based deduplication

Issue Date:2013-01-29

•19 Claims

•10 Drawing Sheets

Abstract

A content alignment system according to certain embodiments aligns a sliding window at the beginning of a data segment. The content alignment system performs a block alignment function on the data within the sliding window. A deduplication block is established if the output of the block alignment function meets a predetermined criteria. At least part of a gap is established if the output of the block alignment function does not meet the predetermined criteria. The predetermined criteria is changed if a threshold number of outputs fail to meet the predetermined criteria.

Metadata

Assignee

CommVault Systems, Inc.

Inventors

Manoj Kumar Vijayan
Deepak Raghunath Attarde
Srikant Viswanathan

Application Information

Application Number:US 12/982,071

Filing Date:2010-12-30

Priority Date:2010-09-30

Art Unit:2169

Classifications

IPC:

G06F17/30

Patent Drawings (10 sheets)

Description

Cross-Reference to Related Application

[0001] This application claims the benefit of priority under 35 U.S.C. §119(e) of U.S. Provisional Patent Application No. 61/388,531, filed on Sep. 30, 2010, and entitled “CONTENT ALIGNED BLOCK-BASED DEDUPLICATION,” the disclosure of which is hereby incorporated by reference in its entirety.

Background

[0002] Computers have become an integral part of business operations such that many banks, insurance companies, brokerage firms, financial service providers, and a variety of other businesses rely on computer networks to store, manipulate, and display information that is constantly subject to change. Oftentimes, the success or failure of an important transaction may turn on the availability of information that is both accurate and current. Accordingly, businesses worldwide recognize the commercial value of their data and seek reliable, cost-effective ways to protect the information stored on their computer networks.

[0003] In corporate environments, protecting information is generally part of a routine process that is performed for many computer systems within an organization. For example, a company might back up critical computing systems related to e-commerce such as databases, file servers, web servers, and so on as part of a daily, weekly, or monthly maintenance schedule. The company may also protect computing systems used by each of its employees, such as those used by an accounting department, marketing department, engineering department, and so forth.

[0004] As such, enterprises are generating ever increasing volumes of data and corresponding storage requirements. Moreover, enterprise storage systems are typically distributed over one or more networks, such as where backup storage is remote from client computers. In such situations, storage system activity can place heavy demands on available network bandwidth.

Summary

[0005] In response to these challenges, one technique developed by storage system providers is data deduplication. Deduplication typically involves eliminating or reducing the amount of redundant data stored and communicated within a storage system, improving storage utilization. For example, a data segment can be divided into units of a chosen granularity (e.g., files or data blocks). As new data segments enter the system, the data units can be checked to see if they already exist in the storage system. If the data unit already exists, instead of storing and/or communicating a duplicate copy, the storage system stores and/or communicates a reference to the existing data unit. Thus, deduplication can improve storage utilization, system traffic (e.g., over a networked storage system), or both.

[0006] One method of dividing the data segment into units is to use fixed-size blocks at fixed-intervals. As an example, a 1 MB data segment can be subdivided into eight 128 kB blocks. Each of the 128 kB blocks is compared to each other as well as other 128 kB blocks within the storage system in order to identify all the identical blocks. However, if a data block differs from previously stored data blocks by just one byte, the result may be no matching data blocks are found, and no data is deduplicated, even though all the data is identical except the one byte.

[0007] In view of the foregoing a need exists to more efficiently determine the appropriate alignment for each block of data, thereby increasing the amount of data that is deduplicated and decreasing storage requirements. According to certain aspects, a sliding window alignment function is performed on data segments to establish block alignments based on the content of the data. The established data blocks can be fixed length, for example. In some embodiments, gaps of data not belonging to any deduplicated blocks exist between the deduplicated data blocks.

[0008] In determining the appropriate alignment some type of predetermined criteria is generally used. Using different criteria in some circumstances can dramatically affect the number of blocks that are deduplicated. Thus, according to certain embodiments, the predetermined criteria is dynamically refined during the deduplication process, providing improved storage utilization.

[0009] In certain embodiments, a method is disclosed for defining deduplication block alignments within a data segment. The method includes iteratively performing a deduplication block alignment function on data within a sliding window in a data segment. For each iterative performance of the deduplication block alignment function, the method further includes in response to determining that the output of the deduplication block alignment function performed on the data within the sliding window satisfies a predetermined criteria establishing with one or more computer processors a deduplication data block having a predetermined block size. For each iterative performance of the deduplication block alignment function, the method further includes in response to determining that the output of the deduplication block alignment function performed on the data within the sliding window does not satisfy the predetermined criteria defining at least a portion of a gap of data not belonging to a deduplication data block.

[0010] In certain embodiments, a system is disclosed for defining deduplication block alignments within a data segment. The system comprises a deduplication block alignment module executing in one or more processors and configured to iteratively perform a deduplication block alignment function on data within a sliding window in a data segment. The block alignment module is further configured for each iterative performance of the block alignment function to establish a deduplication block having a predetermined block size in response to determining that the output of the deduplication block alignment function performed on the data within the sliding window satisfies a predetermined criteria. The block alignment module is further configured for each iterative performance of the block alignment function to define at least a portion of data not belonging to a deduplication block in response to determining that the output of the block alignment function performed on the data within the sliding window does not satisfy the predetermined criteria, for each iterative performance of the block alignment function.

[0011] In certain embodiments, a method is disclosed for defining deduplication block alignments within a data segment. The method includes iteratively performing a block alignment function on data within a sliding window in a data segment. For each iterative performance of the alignment function, the method further includes establishing with one or more computer processors a deduplication data block having a predetermined block size and moving the sliding window relative to the data segment by an amount based on the predetermined block size before performing the next iteration in response to determining that the output of the block alignment function performed on a current window of data satisfies a predetermined criteria. For each iterative performance of the alignment function, the method further includes in response to determining that the output of the block alignment function performed on the current window of data does not satisfy the predetermined criteria moving the sliding window relative to the data segment by an incremental amount before performing the next iteration and without establishing a deduplication data block, wherein gaps of data not belonging to any deduplication data block exist between established deduplication data blocks following performance of the block alignment function across the data segment.

[0012] In certain embodiments, a method is disclosed for determining deduplication block alignments within a data segment. The method includes selecting a first range of possible output values of a deduplication block alignment function which indicate that a block alignment has been found and iteratively performing the deduplication block alignment function on data within a sliding window in a data segment. For each iterative performance of the alignment function, the method further includes determining with one or more computer processors whether the output of the deduplication block alignment function performed on the data within the sliding window falls within the first range. For each iterative performance of the alignment function, the method further includes establishing with one or more computer processors a deduplication data block having a predetermined block size in response to determining that the output of the block alignment function falls within the first range. The method further includes selecting a second range of output values for the block alignment function which indicate that a block alignment has been found, the selection of the second range performed in response to determining, for a threshold number of iterations, that the output of the block alignment does not fall within the first range, wherein the second range is used instead of the first range for subsequent iterations of the block alignment function.

[0013] In certain embodiments, a deduplication system is disclosed for configured determining deduplication block alignments within a data segment. The system includes a block alignment module executing in one or more processors. The block alignment module is configured to select a first range of output values for a deduplication block alignment function which indicate that a block alignment has been found, and iteratively perform the deduplication block alignment function on data within a sliding window in a data segment. The block alignment module is further configured for each iterative performance of the block alignment function to determine whether the output of the deduplication block alignment function performed on data within the sliding window falls within the first range, and establish a deduplication data block with a predetermined block size in response to determining that the output of the deduplication block alignment function performed on the data within the sliding window falls within the first range. The system further includes a criteria adjustment module configured to select a second range of output values for the block alignment function which indicate that a block alignment has been found, the selection of the second range in response to the block alignment module determining, for a threshold number of iterations, that the output of the block alignment performed on the data within the sliding window does fall within the first range, wherein the second range is used instead of the first range for subsequent iterations of the block alignment function.

[0014] In certain embodiments, a method is disclosed for refining criteria for determining deduplication block alignments within a data segment. The method includes selecting a first range of output values for a deduplication block alignment function which indicate that a block alignment has been found, and iteratively performing a block alignment function on data within a sliding window in a data segment. For each iterative performance of the alignment function, the method further includes in response to determining with one or more computer processors whether the output of the block alignment function performed on a current window of data falls within the first range establishing a deduplication data block having a predetermined block size, and moving the sliding window in a first direction relative to the data segment by an amount based on the predetermined block size before performing the next iteration. For each iterative performance of the alignment function, the method further includes in response to determining that the output of the block alignment function performed on the current window of data does not fall within the first range for a threshold number of iterations, selecting a second range of output values for the block alignment function which indicate that a block alignment has been found, and moving the sliding window over data relative to the data segment in a second direction opposite the first direction before performing the next iteration using the second range, wherein the next iteration is performed on data on which the block alignment function was previously performed using the first range.

Description of the Drawings

[0015] FIG. 1 is a block diagram that illustrates components of an example storage system configured to implement techniques compatible with embodiments described herein.

[0016] FIG. 2 is a block diagram illustrative of an example data segment during deduplication in accordance with embodiments described herein.

[0017] FIGS. 3 and 4 are block diagrams illustrative of other examples data segments during deduplication in accordance with embodiments described herein.

[0018] FIG. 5 is a block diagram showing an expanded view of an example data segment including block alignments created in accordance with embodiments described herein.

[0019] FIG. 6 is a flow diagram illustrative of one embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment.

[0020] FIG. 7 is a flow diagram illustrative of one embodiment of a routine implemented by a storage system for deduplicating data blocks within a data segment.

[0021] FIG. 8 is a flow diagram illustrative of another embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment.

[0022] FIG. 9 is a flow diagram illustrative of another embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment.

[0023] FIG. 10 is a flow diagram illustrative of another embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment.

Detailed Description

[0024] Generally described, the present disclosure is directed to a system, method, and computer-readable non-transitory storage medium for defining deduplication block alignments within a data segment and deduplicating data blocks within a data segment. Specifically, aspects of the disclosure will be described with regard to iteratively determining block alignments for a data segment that is to be deduplicated. Although various aspects of the disclosure will be described with regard to examples and embodiments, one skilled in the art will appreciate that the disclosed embodiments and examples should not be construed as limiting.

[0025] As described above, determining the alignment of deduplication blocks based on the content of the data according to techniques described herein can significantly increase the probability that the data will be deduplicated, and that a data segment will use less storage. The present disclosure describes certain embodiments that determine the alignment of deduplication blocks using a block alignment function. A sliding window moves across the data segment and the block alignment function is performed on the data in the sliding window. Deduplication blocks are formed when the output of the block alignment function meets predetermined criteria, which may also be referred to as a “hit.” Gaps, or parts of gaps, are formed when the output of the block alignment function does not meet predetermined criteria i.e. when there is no “hit.”

[0026] After the sliding window has moved across the data segment, the data segment includes deduplication blocks and gaps between some or all of the deduplication blocks. The deduplication blocks can be a fixed-size or variable-size. The gaps between the deduplication blocks will likely vary in size. The size of the gaps are reflective of the number of iterations between the block alignment function returning hits. An example, this will be described in greater detail below with reference to FIG. 5.

[0027] The deduplication blocks are then compared with each other and other previously stored deduplication blocks to determine if they contain identical data. The gaps, however, may not under go this deduplication process. After the deduplication blocks are compared, the deduplication blocks with identical data as other deduplication blocks are replaced with references, often pointers, to the other deduplication blocks. The remaining data segment includes the gaps, references to the deduplication blocks with identical data, and the deduplication blocks that did not have identical data stored elsewhere. Such techniques generally provide improved storage utilization in an efficient manner.

[0028] By dynamically aligning the deduplication blocks, the storage system will increase the likelihood of finding identical data, which can then be replaced with a reference to the other stored data. Thus, the amount of memory required to store the data segment can be significantly reduced, leading to significant cost-savings for enterprises.

[0029] In addition, the description includes embodiments for altering, or tuning, the predetermined criteria during the block alignment phase based on the data. As the sliding window moves across the data, the storage system can determine that there are insufficient hits using the current block alignment function. The storage system can modify the predetermined criteria to increase the likelihood of hits and/or modify the block alignment function. Thus, the storage system can dynamically adapt to the data segment, further leading to an increased probability of finding identical data and reducing the amount of memory used by the data segment. In the present disclosure, a data segment can generally refer to a set of data that includes smaller components. A data segment may be of any size, and may be in the form of stored data or streaming data. A data segment may take the form of parts of a file, an entire file, or multiple files.

[0030] In the present disclosure, a sliding window can generally refer to a boundary used to analyze different parts of a data segment.

[0031] In the present disclosure, a block alignment function can generally refer to any type of function that can be used to determine block alignments. A specific example of a block alignment function is a hash function, particularly weak hash functions.

[0032] In the present disclosure, a block comparison function can generally refer to any type of function capable of comparing two different sets of data within a data segment. A block comparison function may be as simple as comparing each bit, or byte, of data from one set of data to each bit, or byte, of data form another set of data. A more complex type of block comparison function involves comparing hash function values of blocks, particularly those of strong hash functions.

[0033] In the present disclosure, deduplication can generally refer to identifying and removing duplicative data within a data segment. Deduplication may refer to the identifying step, the removing step, or both.

[0034] Illustrative explanations of several terms used throughout the disclosure are provided above. While these meanings apply to the respective terms as used with respect to certain embodiments, it will be appreciated that the meanings can vary depending on the embodiment. Additionally, the meanings of these and other terms used herein will be understood in view of their usage throughout the entirety of the disclosure.

[0035] FIG. 1 illustrates a block diagram of an example network storage architecture compatible with embodiments described herein. The system 100 is configured to perform storage operations on electronic data in a computer network. As shown, the storage system 100 includes a storage manager 102 and one or more of the following: a client 185, an information store 190, a data agent 195, a media agent 105, an index cache 110, and a storage device 115.

[0036] A data agent 195 can be a software module that is generally responsible for archiving, migrating, and recovering data of a client computer 185 stored in an information store 190 or other memory location. Each client computer 185 has at least one data agent 195 and the storage system 100 can support many client computers 185. The storage system 100 provides a plurality of data agents 195 each of which is intended to backup, migrate, and recover data associated with a different application. For example, different individual data agents 195 may be designed to handle Microsoft Exchange™ data, Microsoft Windows file system data, and other types of data known in the art. If a client computer 185 has two or more types of data, one data agent 195 may be implemented for each data type to archive, migrate, and restore the client computer 185 data.

[0037] The storage manager 102 is generally a software module or application that coordinates and controls the system. The storage manager 102 communicates with all elements of the storage system 100 including client computers 185, data agents 195, media agents 105, and storage devices 115, to initiate and manage system backups, migrations, recoveries, and the like.

[0038] A media agent 105 is generally a software module that conducts data, as directed by the storage manager 102, between locations in the storage system 100. For example, the media agent may conduct data between the client computer 185 and one or more storage devices 115, between two or more storage devices 115, etc. The storage devices 115 can include a tape library, a magnetic media storage device, an optical media storage device, or other storage device. Although not shown in FIG. 1, one or more of the media agents 105 may also be communicatively coupled to one another.

[0039] Each of the media agents 105 can be communicatively coupled with and control at least one of the storage devices 115. The media agent 105 generally communicates with the storage device 115 via a local bus. In some embodiments, the storage device 115 is communicatively coupled to the media agent(s) 105 via a Storage Area Network (“SAN”).

[0040] Further embodiments of storage systems such as the one shown in FIG. 1 are described in application Ser. No. 10/818,749, now U.S. Pat. No. 7,246,207, issued Jul. 17, 1007, which is incorporated by reference herein. In various embodiments, components of the storage system 100 may be distributed amongst multiple computers, or one or more of the components may reside and execute on the same computer.

[0041] Furthermore, components of the storage system 100 of FIG. 1 can also communicate with each other via a computer network. For example, the network may comprise a public network such as the Internet, virtual private network (VPN), token ring or TCP/IP based network, wide area network (WAN), local area network (LAN), an intranet network, point-to-point link, a wireless network, cellular network, wireless data transmission system, two-way cable system, interactive kiosk network, satellite network, broadband network, baseband network, combinations of the same or the like.

[0042] Additionally, the various components of storage system 100 may be configured for deduplication. For example, one or more of the clients 185 can include a deduplicated database (DDB). The data stored in the storage devices 115 may also be deduplicated. For example, one or more of the media agents 105 associated with the respective storage devices 115 can manage the deduplication of data in the storage devices 115.

[0043] FIG. 2 is a block diagram illustrative of a data segment 200 during deduplication. The deduplication techniques described with respect to FIG. 2 may be performed by any number of different components of storage system 100, described above with reference to FIG. 1, or some other storage system. For example, deduplication may be performed by the client 185, data agent 195, storage manager 102 any one of the media agents 105. Furthermore, the deduplication may be performed on streaming data being sent to or from the client 185, storage manager 102, and/or media agent 105. Deduplication can also be performed on data residing in any of the client 185, storage manager 102, media agents 105, and/or storage devices 115. In one example scenario, one or more of the storage devices 115 comprise backup storage, and the media agents 105 generally manage deduplication of backup data that is stored on the storage devices 115. Generally, any of the block alignment techniques described herein such as those described with respect to any of FIGS. 2-10 may be performed by a deduplication block alignment module executing on one or more of the components in the system. For example, a block alignment module may be implemented on the storage manager, one or more of the media agents, or a combination thereof.

[0044] As shown in FIG. 2, a data segment 200 may be made up of any number of data subcomponents 202. For ease of reference, each data subcomponent 202 in the example of FIG. 2 represents one byte of information, however, the data subcomponents 202 can take any number of different forms such as bits, bytes or multiples thereof. As discussed previously the data segment 200 may be streaming data or stored in computer memory. The data segment may comprise a file, multiple files or a portion of a file.

[0045] As part of deduplication, a block alignment function is performed on one or more data subcomponents 202 within the data segment 200. A sliding window can be used to determine what data is to be used. The sliding window may be any number of sizes. For example, in various embodiments, the sliding window can be 32 kilobytes (kB), 64 kB, 128 kB, or some smaller or larger value. In the example of FIG. 2, the sliding window covers six subcomponents of data, or six bytes. Thus, there are six bytes of data within the sliding window at any given time. However, it is to be understood that the sliding window may be any size, and the example illustrated in FIG. 2 is not to be construed as limiting. The sliding window may also vary in size as it moves relative to the data segment.

[0046] The block alignment function used can be any number of different functions to determine an appropriate alignment for a deduplication block. In the example of FIG. 2, a hash function is used. In other embodiments, the block alignment function may comprise or output a message digest, checksum, digital fingerprint, digital signature or other sequence of bytes that substantially uniquely identifies the relevant data block. When a hash function is used, the hash function returns a value for the data within the sliding window. As illustrated in FIG. 2, a first hash 204 is performed on the data at the beginning of the data segment 200, which is within the sliding window. The first hash, or hash1, 204 returns a hash value. The hash value is then compared with a predetermined range of values. If the hash value falls within the predetermined range a deduplication block is established, as illustrated at 218. While the example system described with respect to FIG. 2 compares the hash value to a predetermined range, the system may evaluate the hash to determine whether it meets a variety of predetermined criteria, as will be discussed in greater detail herein.

[0047] The size of the deduplication blocks 222, 224, 226, 228 in the illustrated embodiment correspond with the size of the sliding window (e.g., six bytes). If the hash value does not fall within the predetermined range then the sliding window moves relative to the data segment by an incremental value. In this example, the incremental value is representative of one data subcomponent 202, or one byte. However the incremental value can be larger or smaller than one subcomponent. For example, the incremental value could be one bit, byte, or multiples thereof.

[0048] Once the sliding window moves relative to the data segment 200, another hash is performed on the data that is within the sliding window after the move. As illustrated in FIG. 2, the data in the second hash, or hash2, 206 is similar to the data in the first hash 204 except that the second byte of data in the first hash 204 is the first byte of data in the second hash. In addition, the last byte of data in the first hash 204 is the second-to-last byte of data in the second hash 206. In other words, in the example illustrated in FIG. 2 the sliding window moves relative to the data segment to the right by one byte, which represents the incremental value in this example. The second hash returns a hash value which is compared with the predetermined range. Once again, if the hash value falls within the predetermined range, a deduplication block is established 218. If not, the sliding window slides by one incremental value, i.e. one byte. The process is continued until a hash returns a value that falls within the predetermined range, or until the sliding window has moved across the entire data segment 200. Returning a hash value within the predetermined range can also be referred to as a “hit”. A hit occurs in FIG. 2 with the third hash, or hash3, 208.

[0049] In the example illustrated in FIG. 2, the third hash 208 returns a hit within the predetermined range, and block1 222 is established based on the hit. Block1 222 designates the data that returned the hit, and is the same size as the sliding window. However, it is envisioned that the deduplication block size can vary from the sliding window size. In this example the data that returned the hit begins at the third byte and ends at the eighth byte. Thus, block1 222 is made up of six bytes and similarly begins at the third byte and ends at the eighth byte. The example illustrated in FIG. 2 shows that additional hits occurred at the hash5 212, hash(N−1) 214 and hashN 216. From those hits block2 224, block(N−1) 226, and blockN 228 are established, respectively.

[0050] With continued reference to FIG. 2, after the block alignment function has performed on the entire data segment 200, data segment 200 has at least four deduplication blocks: block1 222, block2 224, block(N−1) 226, and blockN 228. In addition to the deduplication blocks, data segment 200 contains gaps between some of the deduplication blocks. The gaps between deduplication blocks represent unsuccessful hashes, or hashes that did not return a hit. For instance, gap 242 is made up of two bytes associated with hash1 and hash2, which did not return hits. Gap 244 is made up of one byte, associated with hash4, which also did not return a hit. Gap 246 represents leftover data that could not become part of a deduplication block after blockN 228 was established because there was insufficient data left to establish another deduplication block.

[0051] As discussed, in other embodiments the deduplication blocks may be larger or smaller than the sliding window. As one illustrative example, the deduplication system performs a hash function on a 64 kB sliding window of data and creates a 128 kB deduplication block. In some embodiments, one or more of the size of the deduplication data blocks and the size of the sliding window may be configurable, either manually by a user, or automatically by the deduplication system itself. For example, the size of the blocks or the size of the sliding window may be adjusted based on a measured deduplication ratio or other appropriate parameter.

[0052] As shown, the system may align the starting point of the deduplication blocks 222, 224, 226, 228 at the beginning of the respective sliding window. For example, the deduplication block 222 begins at the third byte of data, which corresponds to the first byte of the sliding window of data on which the hash3 208 was performed. In other embodiments, the deduplication block alignment starting points can be placed at some other position relative to the sliding window. For example, the system can align the starting point of the deduplication blocks at some other point within the sliding window (e.g., in the middle of the sliding window). In yet other instances, the beginning of the deduplication blocks is defined at some point outside of the sliding window. For example, in one embodiment, the system establishes deduplication blocks such that they begin at a position in the data segment 200 before the respective sliding window. The deduplication blocks in such an embodiment can extend from this starting alignment to include some or all of the data within the sliding window.

[0053] Moreover, the data blocks can extend beyond the sliding window. For example, in the above embodiment where the deduplication system performs a hash function on a 64 kB sliding window of data and creates a 128 kB deduplication block, the deduplication block can extend beyond the sliding window, depending on the starting alignment. For instance, where such a block is aligned at the beginning of the sliding window, the block extends 64 kB beyond the sliding window.

[0054] Once the deduplication blocks for the data segment have been established, the deduplication blocks are deduplicated. To deduplicate the blocks a block comparison function can be performed on the different deduplication blocks. For example, the block comparison function can compare each byte of data in each deduplication block. Alternatively, the block comparison function can be a hash function. Thus, a hash function, different from the hash function used to establish the deduplication blocks, can be used to compare the deduplication blocks. Similar to what is described above, the hash function for each block returns a hash value. The hash value of each deduplication block can be compared. If the hash function of two deduplication blocks returns the same hash value, it is highly probable that the two deduplication blocks contain equivalent data. Thus, the data of one of the deduplication blocks can be removed from the data segment and replaced with a reference to the other deduplication block. In this way, the amount of data stored in the data segment can be reduced. In addition, the hash values of the deduplication blocks of the data segment can be compared with hash values of deduplication blocks from other data segments. While comparing the hash values of the deduplication blocks of the data segment with the hash values of deduplication blocks from other data segments can increase complexity and require additional resources, it can also increase the likelihood of finding equivalent hash values.

[0055] In the example illustrated in FIG. 2, an equivalent hash value is found for the hash value of block1 222, block2 224 and block(N−1) 226. Thus, the data of block1 222, block2 224 and block(N−1) 226 is replaced with references B1 234, B2 236, and BN−1 238. The equivalent hash value can be the same or different for each of block1 222, block2 224 and block(N−1) 226. In addition, no equivalent hash value was found for BlockN 228. Thus, the data of blockN 228 is not replaced with a reference and blockN 228 remains as part of the deduplicated data segment 232. Thus, in the example illustrated in FIG. 2, after the deduplication process, data segment 232 contains gap 242, reference B1 234, gap 244, reference B2 236, reference BN−1 238, blockN 228 and gap 246.

[0056] Although not illustrated in FIG. 2, deduplication blocks for the entire data segment need not be created before the blocks are deduplicated. For example, once hash1 204 returns a hit, block1 222 can be created and immediately deduplicated. In other words, prior to hash4 210, the storage system can compare the hash value of block1 222 with the hash value of other deduplication blocks to determine if block1 222 can be replaced with reference B1 234. In various embodiments, deduplication occurs after creating deduplication block alignments for multiple blocks, an entire segment, multiple segments, a file comprised of multiple segments, or some other granularity of data. Thus, the process for establishing deduplication blocks and deduplicating data segments can occur in any number of ways without departing from the spirit and scope of the disclosure.

[0057] In another embodiment, the storage system moves the sliding window relative to the data segment, and performs the block alignment function for all potential deduplication blocks before determining if the output of the block alignment function for the potential deduplication blocks meets the predetermined criteria. In such an embodiment, the storage system can evaluate the different outputs to determine the criteria to be used as the predetermined criteria. In this way, the storage system can potentially reduce the number of gaps within the deduplication blocks.

[0058] Because the gaps 242, 244, 246 of data are not included in a deduplication block, the gaps 242, 244, 246 generally represent data in the data segment that is not used in the deduplication process. However, the blocks are aligned according to the alignment function based on the content of the data, improving the odds that the deduplication blocks will be redundant. Thus, while all of the data may not be used in the deduplication process, the overall storage utilization is improved according to embodiments described herein. Moreover, in certain embodiments, the deduplication blocks all have the same, fixed length. Using fixed length blocks can reduce the computational overhead involved in creating and comparing deduplication blocks for redundancies and/or simplify the management and storage of the deduplication blocks once created, among providing other advantages.

[0059] FIG. 3 illustrates another example of a data segment 300 during duplication. In the example illustrated in FIG. 3, the block alignments have already been determined i.e. the deduplication blocks and gaps have already been established. In FIG. 3, after the block alignments have been established, the data segment 300 is contains gap 312, block1 304, gap 314, block2 306, gap 316 and blockN 308. Similar to FIG. 2, the gaps between deduplication blocks represent hashes that did not return values within the predetermined range, or outputs of the block alignment function that did not meet the predetermined criteria. Thus, between block1 304 and block2 306 there were several unsuccessful hashes occurred, which resulted in gap 314.

[0060] Also similar to FIG. 2, FIG. 3 shows the data segment 300 after deduplication, represented as deduplicated segment 320. The deduplicated segment 320 is contains gap 312, reference B1 322, gap 314, reference 324, gap 316 and reference BN. As discussed earlier, the references refer to other deduplication blocks that contain data equivalent to the data in the deduplication blocks corresponding to the references. Thus, during deduplication, a second hash function of block1 304, block2 315 and blockN 308 return a hash value equivalent to a hash value of another deduplication block. Thus, block1 304, block2 315 and blockN 308 are replaced with references to the deduplication block with the equivalent hash value. It should be noted that block1 304, block2 315 and blockN 308 may each be equivalent to the same or different deduplication blocks and references B1 322, B2 324, and BN 326 may all refer to different deduplication blocks or the same deduplication block.

[0061] FIG. 4 illustrates another example of a data segment 400 after the block alignments have already been determined. In the example illustrated in FIG. 4, the initial block alignment function resulted in block1 404, block2 406 and blockN 408. A large gap, similar to gap 314 of FIG. 3 is established. However, to reduce the number of gaps, a storage system can determine that once a gap is as large as a deduplication block, a deduplication block should be established from the gap. This determination and establishment of an additional deduplication block is illustrated with blockX 410. In the example illustrated in FIG. 4, the storage system determines that the gap between block1 404 and block2 406 is large enough to allow the establishment of blockX 410, despite that there were no hits between block1 404 and block2 406. To accomplish this, the storage system tracks the number of unsuccessful hits. Once the number of hits reaches some threshold, the storage system can establish the deduplication block.

[0062] In the example illustrated in FIG. 4, the deduplicated data segment 420 contains gap 412, reference B1 422, blockX 410, 414, reference B2 424, gap 416 and reference BN 426. Thus, in the example illustrated in FIG. 4, during the block comparison function, there was not another block that returned a hash value equal to that of blockX 410. Similar to the deduplicated segments illustrated in FIG. 2 and FIG. 3, each of the references B1 422, B2 424 and BN 426 refer to deduplication blocks with data equivalent to that found in block1 404, block2 406 and blockN 408, respectively. In the example illustrated in FIG. 4, the references B1 422, B2 424 and BN 426 can refer to deduplication blocks in other data segments or can point to blockX 410. In another scenario, the system does find a hit for blockX when deduplicating the blocks 418, and the deduplicated segment 420 therefore includes a reference to blockX instead of blockX itself.

[0063] FIG. 4 can also be used to illustrate another function that can be performed on data segment 402. As the sliding window moves relative to the data segment 400, a counter can be used to track the number of times a hash does not return a hit. If a threshold number of hashes fails to return a hit the predetermined range or other criteria can be adjusted. In one embodiment, adjusting the predetermined criteria can include expanding the range of values considered to be a hit. In another embodiment, the block alignment function can be changed entirely. In another embodiment, the sliding window size can be adjusted. To determine how to expand the range of values, or otherwise change the predetermined criteria, the hash values of previous hashes can be examined. Thus, in the example illustrated in FIG. 4, block1 404 returns a hit, but the following several hashes fail to return a hit. The storage system analyzes the hash values of the previous unsuccessful hashes and expands, or otherwise adjusts the range to increase the hits. In the example illustrated in FIG. 4, a threshold of six no-hits is used. In other words, after six iterations without returning a hit, the storage system can adjust the range of values considered to be a hit. In the example illustrated in FIG. 4, after block1 404, six hashes are performed, all of which fail to return a hit. The storage system analyzes the values of the hits and adjusts the range so that at least one additional hit occurs. Adjusting the range results in blockX 410 being established. After adjusting the range of values, the sliding window continues. In one embodiment, after adjusting the predetermined criteria, the storage system can move the sliding window relative to the data segment to the beginning of the data segment to re-analyze the hashes in light of the changed criteria. In another embodiment, after the adjusting the predetermined criteria, the storage system can reevaluate the hash values of the previous hashes and establish any new deduplication blocks in light of the reevaluation before continuing. Additionally, in embodiments where the range of hash values is expanded, the modified range includes the values that were in the original range and there is thus a relatively high probability of finding a matching suitable block boundary.

[0064] Continuing on with the example of FIG. 4, at least two additional hits occur with block2 406 and blockN 408. In another embodiment, once the predetermined criteria is adjusted, the sliding window can return to the beginning of the data segment 400. In an embodiment where the block alignment function is adjusted, the sliding window can begin at the beginning of the data segment. In an alternative embodiment, once the predetermined criteria is adjusted, only some of the data is reevaluated. Depending on the embodiment, the initial and refined predetermined ranges (or other criteria) can differ. As one illustrative example, the block alignment function is a hash function that outputs at 16-bit value having 65,536 possible values. An intermediate function such as a modulo or other operation can be performed on the output values. For example, the intermediate function may reduce the number of values that are analyzed in the block alignment determination. For example, a modulo-100 operation is performed such that each analyzed value is from between 0 and 99. The first predetermined range in the example scenario is from 45 to 55. Where the criteria is adjusted as described herein, the second predetermined range is from 40 to 60. It will be appreciated that the number of output values, the intermediate function, the first range and/or the second range can vary. In other embodiments, the first range and the second range do not overlap or only partially overlap. Moreover, the range may be adjusted more than once in certain embodiments to a third refined range, fourth refined range, etc., or may revert back to previous ranges. Moreover, a modulo function is not used in some cases, or a different intermediate function is used.

[0065] FIG. 5 illustrates a block diagram of an example memory module 502, including data segment 504, after the deduplication block alignments have been determined. The data segment 504 may be any size, but as an example and not to be construed as limiting, the data segment 504 is 1 MB in size. In addition, as an example and not to be construed as limiting, the deduplication block size is a fixed 128 kB. Thus, the data segment 504 could potentially contain a total of eight deduplication blocks. As described above, with reference to FIGS. 2-4, a block alignment function is performed on the data within a sliding window to produce a block alignment value. The sliding window can be the same size as a deduplication block size. As described above, if the block alignment function returns a hit a deduplication block is established and the sliding window moves relative to the data segment according to the size of the deduplication block. If the block alignment function does not return a hit, the sliding window moves relative to the data segment according to the size of an incremental value. As discussed previously, the incremental value may be any number of different sizes ranging from one data subcomponent or more. A block alignment function is performed on the data within the moved sliding window, and the process continues. FIG. 5 illustrates a data segment after this process has occurred across the entire data segment. Data segment 504 contains various deduplication blocks (506, 508, 510, 512, 514, 516, 518) of fixed size. In this example, the fixed size for each deduplication block is 128 kB. However, as mentioned previously, the fixed size may be any size. The data segment also contains various gaps (520, 522, 524, 526, 528) interspersed between the various deduplication blocks (506, 508, 510, 512, 514, 516, 518). The gaps vary in size depending on the number of unsuccessful outputs of the block alignment function. The more unsuccessful outputs from the block alignment function, the larger the gap.

[0066] Reviewing the data segment 504 indicates that there were no hits on the first bytes or other units of data within the sliding window. The first hit occurred at block1 506. No hits occurred between block1 506 and block2 508, resulting in gap 522. The same can be said for gaps 524, 526, and 528. FIG. 5 also illustrates that the block alignment of a data segment 504 will likely result in fewer deduplication blocks being established than could potentially be established. For example, if there was a hit each time a block alignment function was performed on the data within the sliding window, a total of eight would be established. However, usually some amount of data will not return a hit, and thus, there will be fewer deduplication blocks than theoretically possible, with gaps therebetween. In the example illustrated in FIG. 5, only seven deduplication blocks of 128 kB each are established within the 1 MB data segment. Thus, there is 128 kB of gap data.

[0067] FIGS. 6-10 are flow diagrams illustrative of various processes or routines for defining deduplication block alignments within a data segment. FIG. 6 is a flow diagram illustrative of one embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment. FIG. 7 is a flow diagram illustrative of one embodiment of a routine implemented by a storage system for deduplicating data blocks within a data segment. FIG. 8 and FIG. 9 are flow diagrams illustrative of different embodiments of a routine implemented by a storage system for defining deduplication block alignments within a data segment. FIG. 10 is a flow diagram illustrative of another embodiment of a routine implemented by a storage system for defining deduplication block alignments within a data segment. The FIGS. 6-10 will now be described in detail.

[0068] FIG. 6 is a flow diagram illustrative of one embodiment of a routine 600 implemented by a storage system for defining deduplication block alignments within a data segment. The storage system may be any one of the components discuss above with reference to FIG. 1. For instance the storage system can be the client 185, the data agent 195, the storage manager 102, or any one of the media agents 105. One skilled in the relevant art will appreciate that the elements outlined for routine 600 may be implemented by one or many storage systems/components. Accordingly, the following illustrative embodiments should not be construed as limiting. In addition, it is to be understood that the order of the blocks may be changed without affecting the nature or scope of the description.

[0069] At block 602, the routine is initiated. At block 604 a storage system positions a sliding window at the beginning of a data segment. However, it is to be noted that the sliding window can be placed at any location before, within, or after the data segment. Thus, the initial location of the sliding window before, within, or after the data segment may also be referred to as the beginning of the data segment. As mentioned above, the data segment can be a stream of data or can be data that is stored in computer memory. The computer memory can be any of RAM, ROM, solid-state, tapes, hard-discs, magnetic storage disks, or the like. The storage system can select the beginning of the data segment based on predetermined criteria or may select the beginning dynamically.

[0070] Once the sliding window has been positioned at the beginning of the data segment to be analyzed, the storage system performs a block alignment function on the data within the sliding window, as illustrated at block 606. As described above, the block alignment function may be any number of functions that allows the storage system to determine an appropriate alignment. Hash functions may provide one alternative for the block alignment function. The hash function may be either a weak or strong hash function. The type of hash function used can be determined based on the amount of computing resources available, time sensitive nature of the deduplication routine, and the like. As is known in the art, weak hash functions typically require less resources than strong hash functions. Thus, in determining the boundary alignment, a weak hash function may be preferred.

[0071] With continued reference to FIG. 6, at decision block 610, the storage system determines if the output of the block alignment function satisfies a predetermined criteria. In one embodiment, the predetermined criteria can be a range of values. For instance, if a hash function is used, a range of values from the potential output values of the hash function may be selected. After the hash function, or other block alignment function, has been performed on the data within the sliding window, if the output of the block alignment function is a value within the range selected, the storage system may determine that the predetermined criteria is met. The range of values may vary depending on the data segment being analyzed and the block alignment function being used.

[0072] If the predetermined criteria is met the storage system establishes a deduplication block, as illustrated in block 610. Establishing a deduplication block may include storing any information that would aid in locating the deduplication block. For instance, in one embodiment, the storage system can store the location of the sliding window, the size of the deduplication block, a pointer to or address of the deduplication block, as well as the output of the block alignment function. In another embodiment, the storage system can store a reference to the deduplication block and the size of the deduplication block may be predetermined. In yet another embodiment, the storage system can store the entire deduplication block for later use. Any number of methods may be implemented that allows the storage system to access the data within the established deduplication block without departing from the spirit and scope of the description.

[0073] Upon establishing the deduplication block, the storage system moves the sliding window relative to the data segment based on the size of the deduplication block, as illustrated in block 612. As noted earlier, the size of the deduplication block may be predetermined or may be determined dynamically. Moving the sliding window can be accomplished in a number of different ways depending on the data. In one embodiment, the sliding window is a buffer. In this embodiment, moving the sliding window includes changing the data within the buffer with new data to be analyzed. In some embodiments, all of the data within the buffer is replaced, in other embodiments some of the data remains. In another embodiment, the sliding window is a pointer or set of pointers to one or more address locations within the data segment. In this embodiment, moving the sliding window can include changing the address location pointed to by the pointer. In an embodiment where the data is streaming, moving the sliding window can be as simple as analyzing the data currently passing the system. In addition, based on the starting point of the sliding window, the sliding window can move forwards or backwards relative to the data segment.

[0074] Furthermore, the sliding window can move in increments larger or smaller than the size of the deduplication blocks. For example, moving the sliding window based on the size of the deduplication block may result in gaps between the newly created deduplication block and the next data within the sliding window, or some data may have the block alignment function performed more than once. If the predetermined criteria is not met, then the storage system moves the sliding window relative to the data segment by an incremental amount, as illustrated at block 614. As mentioned above, the incremental amount may be any one of many different amounts. For example, the incremental amount maybe a bit, a byte, several bytes, hundreds, thousands, or even more bytes depending on the data segment.

[0075] Once the sliding window moves relative to the data segment as discussed with reference to either block 612 or 614, the storage system determines if there is more data in the data segment, as illustrated in block 616. If there is no more data in the data segment, the storage system can end the process, as illustrated in block 618. As part of this determination, the storage system can determine if there is sufficient data left with which to make a deduplication block. If there is insufficient data remaining in the segment to make a deduplication block, the storage system can determine that the end of the segment has been reached, even though some data remains, and end the routine, as illustrated in block 618.

[0076] If, however, the storage system determines that additional data remains within the data segment, the storage system can continue determining the block alignment for the additional data and perform the block alignment function on the data within the moved sliding window, as illustrated in block 606. In addition, as part of the end of segment determination in determining block 616 if storage system determines that there is sufficient data remaining in the data block to establish only one more data block without any intermediate gaps or gaps at the end of the data block, the storage system can establish a deduplication block without performing a block alignment function.

[0077] With further reference to FIG. 6, it is to be understood that the order of the blocks may be changed without affecting the nature or scope of the description, or the routine 600 may be performed with more blocks or fewer blocks. For example, although not illustrated, routine 600 may store the output of the block alignment function as well as the location of the alignment window.

[0078] FIG. 7 is a flow diagram illustrative of one embodiment of a routine 700 implemented by a storage system for deduplicating data blocks within a data segment. One skilled in the relevant art will appreciate that the elements outlined for routine 700 may be implemented by one or many storage systems/components. Accordingly, the following illustrative embodiments should not be construed as limiting.

[0079] At block 702, the routine is initiated. At block 704, the storage system performs a block comparison function on a first data block. The first data block may be the first established data block from block 610 of FIG. 6. In an embodiment, routine 700 follows routine 600. In such an embodiment, storage system determines block alignments and established data blocks using routine 600, and uses routine 700 to deduplicate the blocks established by routine 600. The block comparison function may be any number of various functions. In one embodiment, the block comparison function may compare each subcomponent of data in two data blocks to determine if the current deduplication data block is equivalent to another, previously stored deduplication data block. For example, the block comparison function may compare the hash of the current deduplication data block to the hashes of previously stored deduplication data blocks. In one embodiment the hash function used may be different from the hash function used in routine 600. For example, in order to reduce or eliminate collisions in which non-redundant data is removed, the hash function can be a strong hash function. In one embodiment, the block comparison function compares the data blocks within the data segment to other data blocks within the data segment. In other embodiments, the block comparison function compares data blocks in one data segment with data blocks in other data segments. Any number of various functions may be used to determine if a data block is equivalent or substantially equivalent to another data block without departing from the spirit and scope of the description.

[0080] Once the block comparison function is performed on the data block, the storage system may determine if the output of the block comparison function meets a predetermined criteria. Generally, the predetermined criteria can be whether the output block comparison function indicates that current deduplication block matches another, previously stored deduplication data block, such as where the output is equal to the output for another, previously stored deduplication data block. The value of the output of the block comparison function on the other data block can be stored in computer memory and can be from the same data segment or from a different data segment. For example, the system in one embodiment maintains a table of hash values corresponding to previously stored deduplication data blocks. If the storage system determines that the output of the block comparison function meets the predetermined criteria (e.g., the hash matches that of another stored data block), the storage system replaces the data block with a reference to the other data block. In this manner, the storage system can reduce the overall size of the data segment. Rather than containing all the data from the data block, the data segment will contain a reference to an equivalent data block.

[0081] If the output of the block comparison function does not meet the predetermined criteria the storage system can determine if there are additional data blocks, as illustrated in block 710. In addition, after the storage system has replaced a data block with a reference, as illustrated in block 708, the storage system will then determine if there are additional data blocks, as illustrated in block 710. If there are additional data blocks, the storage system can perform the block comparison function on the next data block, as illustrated in block 712. If, however, there are no more data blocks, then the storage system can end the routine 700, as illustrated in block 714.

[0082] With further reference to FIG. 7, it is to be understood that the order of the blocks may be changed without affecting the nature or scope of the description, or the routine 700 may be performed with more blocks or fewer blocks. For example, although not illustrated routine 700 may store the output of the block comparison function.

[0083] FIG. 8 and FIG. 9 are flow diagrams illustrative of alternative embodiments of routines 800 and 900 implemented by a storage system for defining deduplication block alignments within a data segment. One skilled in the relevant art will appreciate that the elements outlined for routines 800 and 900 may be implemented by one or many storage systems/components. Accordingly, the following illustrative embodiments should not be construed as limiting.

[0084] FIG. 8 and FIG. 9 further illustrate alternative embodiments for altering, or refining, the predetermined criteria during deduplication. As mentioned above, refining the criteria as the sliding window moves across the data segment can further increase the likelihood of finding identical data and reducing the amount of memory used by the data segment. At a high-level, the process 800 refines the criteria and then reevaluates at least some of the data within the data segment. This may occur, for instance, when refining the criteria includes changing the block alignment function. The process 900, on the other hand, refines the criteria and continues evaluating the data segment without reevaluating previously evaluated data. This may occur, for example, when refining the criteria includes expanding, or changing, the range considered to be a hit. Generally, any of the alignment criteria adjustment techniques described herein such as those described with respect to any of FIGS. 4, 8 and 9 may be performed by a criteria adjustment module executing on one or more of the components in the system. For example, a criteria adjustment module may be implemented on the storage manager, one or more of the media agents, or a combination thereof.

[0085] Blocks 802-812 and blocks 902-912 of FIG. 8 and FIG. 9, respectively, are similar to and correspond with blocks 602-612 of FIG. 6. At block 802 and block 902 the storage system, positions the sliding window at the beginning of a data segment. At blocks 804 and 904, the storage system performs a block alignment function on the data within the sliding window. At decision blocks 808 and 908, the storage system determines if the output of the block alignment function satisfies a predetermined criteria. If the output of the block alignment function satisfies a predetermined criteria the storage system establishes a deduplication block, as illustrated in blocks 810 and 910, and moves the sliding window relative to the data segment based on the size of the deduplication block, as illustrated in blocks 812 and 912.

[0086] In blocks 814 and 914, the storage system resets a gap counter, which will be explained in greater detail below. Blocks 816-818 of FIG. 8 and blocks 916-918 of FIG. 9 correspond with block 616-618 of FIG. 6. In blocks 816 and 916, the storage system determines if the end of the segment has been reached. Similar to block 616 of FIG. 6, the storage system may make this determination in any number of different ways. If the storage system determines that the end of the data segment has been reached, the storage system ends the routine, as illustrated in blocks 818 and 918.

[0087] Referring now to only FIG. 8, if the storage system determines at determination block 808 that the output of the block alignment function does not satisfy a predetermined criteria, the storage system increments a gap counter, as illustrated in block 820. The gap counter tracks the number of consecutive instances that the data in the sliding window has failed to satisfy the predetermined criteria. As mentioned previously, when the output of the block alignment function satisfies a predetermined criteria the gap counter is reset. Thus, by incrementing each time the output of the block alignment function fails to satisfy the predetermined criteria, the size of the gap between each deduplication block can be monitored. The gap counter may be implemented in hardware or software in any number of ways without departing from the spirit and scope of the description.

[0088] With continued reference to FIG. 8, the storage system determines if a threshold number of no-hits has been reached, as illustrated in block 822. In some embodiments, if there have not been enough hits, i.e. there have not been enough outputs from the block alignment function that satisfy the predetermined criteria, it may be beneficial to adjust the predetermined criteria to increase the number of hits within a data segment. Thus, a threshold number of gaps, or no-hits, can indicate when the predetermined criteria should be adjusted.

[0089] If at decision block 822, the storage system determines that the gap counter has not reached a threshold number, the storage system moves the sliding window relative to the data segment based on an incremental value, as illustrated in block 824, similar to block 614 of FIG. 6.

[0090] On the other hand, if at determination block 822, the storage system determines that the threshold has been reached, the storage system may adjust the predetermined criteria and reset the gap counter, as illustrated in blocks 826 and 828, respectively. In adjusting the predetermined criteria, the storage system may perform any number of tasks, such as increase the range of acceptable values, establish a new range of acceptable values different from the previous range, or can adjust the block alignment function. In one embodiment, the storage system may adjust the block alignment function by using a different hash function. In another embodiment, the storage system can adjust the size of the sliding window.

[0091] Upon adjusting the predetermined criteria, as illustrated in block 826, and resetting the gap counter, as illustrated in block 828, the storage system may then position the sliding window at the beginning of the data segment and begin computing the block alignment function from the beginning, as illustrated in block 804. Although not illustrated in FIG. 8, in some embodiments only some of the data within the data segment is reevaluated. For instance, the storage system moves the sliding window to an earlier location within the data segment that is not the beginning of the data segment, or the initial location of the sliding window. An alternative embodiment is also described with reference to FIG. 9, where the sliding window moves based on an incremental value after the criteria is refined.

[0092] Blocks 920 and 922 of FIG. 9 correspond with blocks 820 and 822 of FIG. 8. At block 920, the gap counter is incremented. At block 922, the storage system determines if the gap counter has reached a threshold number, similar to block 822 of FIG. 8. Also similar to FIG. 8, if the gap counter has not reached the threshold value, the storage system moves the sliding window relative to the data segment according to an incremental value, as illustrated in block 924 and makes the end of segment determination, as illustrated in block 916, and described in greater detail above.

[0093] However, if the storage system determines that the gap counter has reached a threshold value, the storage system adjusts the predetermined criteria, as illustrated in block 926. Adjusting the predetermined criteria may be done in a similar manner as described earlier with reference to block 826 of FIG. 8. The storage system can also reset the gap counter, as illustrated in block 928, and described in greater detail above, with reference to block 828 of FIG. 8.

[0094] With further reference to FIG. 8 and FIG. 9, it is to be understood that the order of the blocks may be changed without affecting the nature or scope of the description. In addition, routines 800 and 900 may be performed with more blocks or fewer blocks. For example, although not illustrated routines 800 and 900 may store the output of the block alignment function. In addition, routines 800 and 900 may reset the gap counter before establishing a deduplication block, moving the sliding window, or adjusting the predetermined criteria. Thus, there are various ways of implementing routines 800 and 900 without departing from the spirit and scope of the description.

[0095] FIG. 10 is a flow diagram illustrative of another embodiment of a routine 1000 implemented by a storage system for defining deduplication block alignments within a data segment. At block 1002, the routine is initiated. At block 1004, the storage system determines if the gap size is equal to a threshold size. As discussed in greater detail above, the gap counter is able to track the number of no-hits and size of the gap between deduplication blocks. In some instances, the size of the gap between deduplication blocks may be the same as, or larger than a deduplication block. In such a scenario, it can be beneficial to create a deduplication block out of the gap. Thus, at block 1004, the storage system determines if the gap size is equal to or larger than a deduplication block. If the storage system determines that the gap size is not equal to or larger than a deduplication block, the storage system moves the sliding window relative to the data segment based on an incremental value, as illustrated in block 1014, and described in greater detail above with reference to block 614 of FIG. 6.

[0096] However, if the storage system determines that the gap size is equal to or greater than a deduplication block, the storage system may establish a deduplication block out of the data in the gap, as illustrated in block 1006. Although not illustrated in FIG. 10, the storage system may also establish a deduplication block out of a portion, or portions, of the data in the gap. The storage system can establish the deduplication block in a manner similar to that described above with reference to block 610 of FIG. 6. The storage system moves the sliding window relative to the data segment based on the size of the newly established deduplication block, as illustrated in block 1008, and described in greater detail above with reference to block 612 of FIG. 6. Although not illustrated in FIG. 10, in some embodiments, the sliding window may not be incremented after a block is created out of the data, or portions thereof. For instance, if a block is created once a gap is the size of a deduplication block, the sliding window may already be positioned in the appropriate location, i.e. over the data that comes after the data in the newly created deduplication block. In other embodiments, the sliding window may be incremented by an incremental value. At block 1010, the storage system can reset the gap counter, as described in greater detail above with reference to block 814 of FIG. 8. At block 1012, the storage system can stop the routine 1000.

[0097] It is to be understood that routine 1000 can be implemented independently or as part of any of routines 600, 700, 800, and/or 900. As such, the determination of whether a deduplication block should be established from a gap of data can be done along with the other determinations made in routines 600, 700, 800, and 900. For example, routine 1000 may be inserted in the place of block 614 of FIG. 6, or in other locations within any of routines 700, 800, or 900.

[0098] With further reference to FIG. 10, it is to be understood that the order of the blocks may be changed without affecting the nature or scope of the description, or the routine 1000 may be performed with more blocks or fewer blocks. In addition, the blocks of routine 1000 may be performed in parallel or interleaved with the blocks of routines 600, 700, 800, or 900. Thus, there are many ways of implementing routine 1000 without departing from the spirit and scope of the description.

[0099] It will be appreciated by those skilled in the art and others that all of the functions described in this disclosure may be embodied in software executed by one or more processors of the disclosed components and mobile communication devices. The software may be persistently stored in any type of non-volatile storage.

[0100] Conditional language, such as, among others, “can,” “could,” “might,” or “may,” unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular embodiment.

[0101] In certain embodiments of the invention, operations disclosed herein can be used to copy or otherwise retrieve data of one or more applications residing on and/or being executed by a computing device. For instance, the applications may comprise software applications that interact with a user to process data and may include, for example, database applications (e.g., SQL applications), word processors, spreadsheets, financial applications, management applications, e-commerce applications, browsers, combinations of the same or the like. For example, in certain embodiments, the applications may comprise one or more of the following: MICROSOFT EXCHANGE, MICROSOFT SHAREPOINT, MICROSOFT SQL SERVER, ORACLE, MICROSOFT WORD and LOTUS NOTES.

[0102] Moreover, in certain embodiments of the invention, data backup systems and methods may be used in a modular storage management system, embodiments of which are described in more detail in U.S. Pat. No. 7,035,880, issued Apr. 5, 2006, and U.S. Pat. No. 6,542,972, issued Jan. 30, 2001, each of which is hereby incorporated herein by reference in its entirety. For example, the disclosed backup systems may be part of one or more storage operation cells that includes combinations of hardware and software components directed to performing storage operations on electronic data. Exemplary storage operation cells usable with embodiments of the invention include CommCells as embodied in the QNet storage management system and the QiNetix storage management system by CommVault Systems, Inc., and as further described in U.S. Pat. No. 7,454,569, issued Nov. 18, 2008, which is hereby incorporated herein by reference in its entirety.

[0103] Systems and modules described herein may comprise software, firmware, hardware, or any combination(s) of software, firmware, or hardware suitable for the purposes described herein. Software and other modules may reside on servers, workstations, personal computers, computerized tablets, PDAs, and other devices suitable for the purposes described herein. Software and other modules may be accessible via local memory, via a network, via a browser, or via other means suitable for the purposes described herein. Data structures described herein may comprise computer files, variables, programming arrays, programming structures, or any electronic information storage schemes or methods, or any combinations thereof, suitable for the purposes described herein. User interface elements described herein may comprise elements from graphical user interfaces, command line interfaces, and other interfaces suitable for the purposes described herein.

[0104] Embodiments of the invention are also described above with reference to flow chart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products. 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, may 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 acts specified in the flow chart and/or block diagram block or blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the acts specified in the flow chart and/or block diagram block or blocks. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operations 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 steps for implementing the acts specified in the flow chart and/or block diagram block or blocks.

[0106] While certain embodiments of the inventions have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the disclosure. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the disclosure. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the disclosure.

Claims

What is claimed is:

1. A method for refining criteria for determining deduplication block alignments within a data segment, the method comprising:

selecting a first range of output values for a deduplication block alignment function which indicate that a block alignment has been found; and

iteratively performing the block alignment function using one or more computer processors on data within a sliding window in the data segment, wherein the sliding window comprises a sliding boundary, and for each iterative performance of the block alignment function:

in response to determining with the one or more computer processors that an output of the block alignment function performed on a current window of data of the data segment falls within the first range:

establishing a deduplication data block having a predetermined block size;

moving the sliding window in a first direction relative to the data segment by an amount based on the predetermined block size before performing a next iteration; and

in response to determining that the output of the block alignment function performed on the current window of data does not fall within the first range for a threshold number of iterations:

selecting a second range of output values for the block alignment function which indicate that a block alignment has been found; and

moving the sliding window relative to the data segment in a second direction opposite to the first direction before performing the next iteration using the second range, wherein the next iteration is performed on data on which the block alignment function was previously performed using the first range.

2. The method of claim 1, wherein the threshold number of iterations is a threshold number of consecutive iterations.

3. The method of claim 1, further comprising, after the second range of output values is selected, establishing with the one or more computer processors one or more deduplication data blocks each having the predetermined block size in response to determining that the output of the block alignment function performed on the data previously within the sliding window falls within the second range.

4. The method of claim 3, wherein the data previously within the sliding window comprises data that did not fall within the first range.

5. The method of claim 1, further comprising moving the sliding window relative to the data segment to a previous location within the data segment in response to determining, for a threshold number of consecutive iterations, that the output of the block alignment function performed on the data within the sliding window does not fall within the first range.

6. The method of claim 5, wherein the previous location is the beginning of the data segment.

7. The method of claim 1, further comprising iteratively performing the block alignment function across at least a portion of the data segment again after selecting the second range.

8. The method of claim 1, wherein the first range overlaps with the second range.

9. The method of claim 8, wherein the first range is a subset of the second range.

10. The method of claim 1, wherein the first range and the second range do not overlap.

11. The method of claim 1, wherein the block alignment function is a hash function and the method further comprises performing an intermediate modulo operation on the output of the block alignment function, and wherein the determining comprises determining whether the output of the intermediate modulo operation falls within the first range.

12. The method of claim 1, further comprising defining at least a portion of a gap of data not belonging to the deduplication data block in response to determining that the output of the block alignment function performed on the data within the sliding window does not fall within the first range.

13. The method of claim 1, further comprising establishing the deduplication data block in response to determining for a second threshold number of consecutive iterations that the output of the block alignment function performed on the current window of data does not fall within the first range.

14. A deduplication system for determining deduplication block alignments within a data segment, the system comprising:

one or more computer processors;

a block alignment module being executed by the one or more computer processors and configured to:

select a first range of output values for a deduplication block alignment function which indicates that a block alignment has been found; and

iteratively perform the block alignment function using one or more computer processors on data within a sliding window in the data segment, wherein the sliding window comprises a sliding boundary, and for each iterative performance of the block alignment function, the block alignment module is configured to:

determine whether an output of the block alignment function performed on a current window of data falls within the first range,

establish a deduplication data block having a predetermined block size in response to determining that the output of the block alignment function performed on the current window of data falls within the first range;

move the sliding window in a first direction relative to the data segment by an amount based on the predetermined block size before performing a next iteration; and

a criteria adjustment module configured to:

select a second range of output values for the block alignment function which indicates that a block alignment has been found in response to determining that the output of the block alignment function performed on the current window of data does not fall within the first range for a threshold number of iterations; and

move the sliding window relative to the data segment in a second direction opposite to the first direction before performing the next iteration using the second range, wherein the next iteration is performed on data on which the block alignment function was previously performed using the first range.

15. The deduplication system of claim 14, wherein the block alignment module is further configured to iteratively perform the block alignment function across at least a portion of the data segment again after selecting the second range.

16. The deduplication system of claim 14, wherein the first range overlaps with the second range.

17. The deduplication system of claim 16, wherein the first range is a subset of the second range.

18. The deduplication system of claim 14, wherein the first range and the second range do not overlap.

19. The deduplication system of claim 14, wherein the block alignment module is further configured to establish the deduplication data block in response to determining for a second threshold number of iterations that the output of the block alignment function performed on the current window of data does not fall within the first range.

Patent Citations (155)

Patent	Date	Inventor	Cited By
US4686620(A)	1987-08-01	Ng	Applicant
US4995035(A)	1991-02-01	Cole et al.	Applicant
US5005122(A)	1991-04-01	Griffin et al.	Applicant
US5093912(A)	1992-03-01	Dong et al.	Applicant
US5133065(A)	1992-07-01	Cheffetz et al.	Applicant
US5193154(A)	1993-03-01	Kitajima et al.	Applicant
US5212772(A)	1993-05-01	Masters	Applicant
US5226157(A)	1993-07-01	Nakano et al.	Applicant
US5239647(A)	1993-08-01	Anglin et al.	Applicant
US5241668(A)	1993-08-01	Eastridge et al.	Applicant
US5241670(A)	1993-08-01	Eastridge et al.	Applicant
US5276860(A)	1994-01-01	Fortier et al.	Applicant
US5276867(A)	1994-01-01	Kenley et al.	Applicant
US5287500(A)	1994-02-01	Stoppani, Jr.	Applicant
US5321816(A)	1994-06-01	Rogan et al.	Applicant
US5333315(A)	1994-07-01	Saether et al.	Applicant
US5347653(A)	1994-09-01	Flynn et al.	Applicant
US5410700(A)	1995-04-01	Fecteau et al.	Applicant
US5448724(A)	1995-09-01	Hayashi et al.	Applicant
US5491810(A)	1996-02-01	Allen	Applicant
US5495607(A)	1996-02-01	Pisello et al.	Applicant
US5504873(A)	1996-04-01	Martin et al.	Applicant
US5544345(A)	1996-08-01	Carpenter et al.	Applicant
US5544347(A)	1996-08-01	Yanai et al.	Applicant
US5559957(A)	1996-09-01	Balk	Applicant
US5619644(A)	1997-04-01	Crockett et al.	Applicant
US5638509(A)	1997-06-01	Dunphy et al.	Applicant
US5673381(A)	1997-09-01	Huai et al.	Applicant
US5699361(A)	1997-12-01	Ding et al.	Applicant
US5729743(A)	1998-03-01	Squibb	Applicant
US5751997(A)	1998-05-01	Kullick et al.	Applicant
US5758359(A)	1998-05-01	Saxon	Applicant
US5761677(A)	1998-06-01	Senator et al.	Applicant
US5764972(A)	1998-06-01	Crouse et al.	Applicant
US5778395(A)	1998-07-01	Whiting et al.	Applicant
US5812398(A)	1998-09-01	Nielsen	Applicant
US5813008(A)	1998-09-01	Benson et al.	Applicant
US5813009(A)	1998-09-01	Johnson et al.	Applicant
US5813017(A)	1998-09-01	Morris	Applicant
US5875478(A)	1999-02-01	Blumenau	Applicant
US5887134(A)	1999-03-01	Ebrahim	Applicant
US5901327(A)	1999-05-01	Ofek	Applicant
US5924102(A)	1999-07-01	Perks	Applicant
US5940833(A)	1999-08-01	Benson	Applicant
US5950205(A)	1999-09-01	Aviani, Jr.	Applicant
US5974563(A)	1999-10-01	Beeler, Jr.	Applicant
US5990810(A)	1999-11-01	Williams	Applicant
US6021415(A)	2000-02-01	Cannon et al.	Applicant
US6026414(A)	2000-02-01	Anglin	Applicant
US6052735(A)	2000-04-01	Ulrich et al.	Applicant
US6076148(A)	2000-06-01	Kedem	Applicant
US6094416(A)	2000-07-01	Ying	Applicant
US6131095(A)	2000-10-01	Low et al.	Applicant
US6131190(A)	2000-10-01	Sidwell	Applicant
US6148412(A)	2000-11-01	Cannon et al.	Applicant
US6154787(A)	2000-11-01	Urevig et al.	Applicant
US6161111(A)	2000-12-01	Mutalik et al.	Applicant
US6167402(A)	2000-12-01	Yeager	Applicant
US6212512(B1)	2001-04-01	Barney et al.	Applicant
US6260069(B1)	2001-07-01	Anglin	Applicant
US6269431(B1)	2001-07-01	Dunham	Applicant
US6275953(B1)	2001-08-01	Vahalia et al.	Applicant
US6301592(B1)	2001-10-01	Aoyama et al.	Applicant
US6324581(B1)	2001-11-01	Xu et al.	Applicant
US6328766(B1)	2001-12-01	Long	Applicant
US6330570(B1)	2001-12-01	Crighton	Applicant
US6330642(B1)	2001-12-01	Carteau	Applicant
US6343324(B1)	2002-01-01	Hubis et al.	Applicant
USRE37601(E)	2002-03-01	Eastridge et al.	Applicant
US6356801(B1)	2002-03-01	Goodman et al.	Applicant
US6389432(B1)	2002-05-01	Pothapragada et al.	Applicant
US6421711(B1)	2002-07-01	Blumenau et al.	Applicant
US6487561(B1)	2002-11-01	Ofek et al.	Applicant
US6519679(B2)	2003-02-01	Devireddy et al.	Applicant
US6538669(B1)	2003-03-01	Lagueux, Jr. et al.	Applicant
US6564228(B1)	2003-05-01	O'Connor	Applicant
US6658526(B2)	2003-12-01	Nguyen et al.	Applicant
US6704730(B2)	2004-03-01	Moulton et al.	Applicant
US6757794(B2)	2004-06-01	Cabrera et al.	Applicant
US6810398(B2)	2004-10-01	Moulton	Applicant
US6865655(B1)	2005-03-01	Andersen	Applicant
US6952758(B2)	2005-10-01	Chron et al.	Applicant
US7085904(B2)	2006-08-01	Mizuno et al.	Applicant
US7117246(B2)	2006-10-01	Christenson et al.	Applicant
US7143091(B2)	2006-11-01	Charnock	Applicant
US7246272(B2)	2007-07-01	Cabezas et al.	Applicant
US7272606(B2)	2007-09-01	Borthakur et al.	Applicant
US7287252(B2)	2007-10-01	Bussiere et al.	Applicant
US7383462(B2)	2008-06-01	Osaki et al.	Applicant
US7437388(B1)	2008-10-01	DeVos	Applicant
US7490207(B2)	2009-02-01	Amarendran	Applicant
US7516186(B1)	2009-04-01	Borghetti et al.	Applicant
US7519726(B2)	2009-04-01	Palliyll et al.	Applicant
US7568080(B2)	2009-07-01	Prahlad et al.	Applicant
US7613748(B2)	2009-11-01	Brockway et al.	Applicant
US7631194(B2)	2009-12-01	Wahlert et al.	Applicant
US7657550(B2)	2010-02-01	Prahlad et al.	Applicant
US7685126(B2)	2010-03-01	Patel et al.	Applicant
US7761425(B1)	2010-07-01	Erickson et al.	Applicant
US7797279(B1)	2010-09-01	Starling et al.	Applicant
US7827150(B1)	2010-11-01	Wu et al.	Applicant
US7831795(B2)	2010-11-01	Prahlad et al.	Applicant
US8074043(B1)	2011-12-01	Zeis	Applicant
US2002/0107877(A1)	2002-08-01	Whiting et al.	Applicant
US2002/0169934(A1)	2002-11-01	Krapp et al.	Applicant
US2003/0033308(A1)	2003-02-01	Patel et al.	Applicant
US2004/0148306(A1)	2004-07-01	Moulton et al.	Applicant
US2005/0033756(A1)	2005-02-01	Kottomtharayil et al.	Applicant
US2006/0010227(A1)	2006-01-01	Atluri	Applicant
US2006/0064456(A1)	2006-03-01	Kalthoff et al.	Applicant
US2006/0095527(A1)	2006-05-01	Malik	Applicant
US2006/0101096(A1)	2006-05-01	Fuerst	Applicant
US2006/0224846(A1)	2006-10-01	Amarendran et al.	Applicant
US2007/0179995(A1)	2007-08-01	Prahlad et al.	Applicant
US2007/0203937(A1)	2007-08-01	Prahlad et al.	Applicant
US2007/0250670(A1)	2007-10-01	Fineberg et al.	Applicant
US2007/0255758(A1)	2007-11-01	Zheng et al.	Examiner
US2008/0091655(A1)	2008-04-01	Gokhale et al.	Applicant
US2008/0133561(A1)	2008-06-01	Dubnicki et al.	Applicant
US2008/0159331(A1)	2008-07-01	Mace et al.	Applicant
US2008/0229037(A1)	2008-09-01	Bunte et al.	Applicant
US2008/0243879(A1)	2008-10-01	Gokhale et al.	Applicant
US2008/0243914(A1)	2008-10-01	Prahlad et al.	Applicant
US2008/0243957(A1)	2008-10-01	Prahlad et al.	Applicant
US2008/0243958(A1)	2008-10-01	Prahlad et al.	Applicant
US2008/0250204(A1)	2008-10-01	Kavuri et al.	Applicant
US2009/0013129(A1)	2009-01-01	Bondurant	Applicant
US2009/0043767(A1)	2009-02-01	Joshi et al.	Applicant
US2009/0055425(A1)	2009-02-01	Evans et al.	Applicant
US2009/0204636(A1)	2009-08-01	Li et al.	Examiner
US2009/0254507(A1)	2009-10-01	Hosoya et al.	Applicant
US2009/0271454(A1)	2009-10-01	Anglin et al.	Applicant
US2009/0319534(A1)	2009-12-01	Gokhale	Applicant
US2009/0319585(A1)	2009-12-01	Gokhale	Applicant
US2010/0005259(A1)	2010-01-01	Prahlad	Applicant
US2010/0036887(A1)	2010-02-01	Anglin et al.	Applicant
US2010/0077161(A1)	2010-03-01	Stoakes et al.	Applicant
US2010/0180075(A1)	2010-07-01	McCloskey et al.	Examiner
US2010/0250896(A1)	2010-09-01	Matze	Examiner
US2010/0299490(A1)	2010-11-01	Attarde et al.	Examiner
US2011/0113016(A1)	2011-05-01	Gruhl et al.	Examiner
US2011/0276543(A1)	2011-11-01	Matze	Applicant
US2011/0289281(A1)	2011-11-01	Spackman	Examiner
US2012/0084518(A1)	2012-04-01	Vijayan et al.	Applicant
US2012/0084519(A1)	2012-04-01	Vijayan et al.	Applicant
EP259912	1988-03-01		Applicant
EP405926	1991-01-01		Applicant
EP467546	1992-01-01		Applicant
EP774715	1997-05-01		Applicant
EP809184	1997-11-01		Applicant
EP899662	1999-03-01		Applicant
EP981090	2000-02-01		Applicant
WO95/13580	1995-05-01		Applicant
WO99/12098	1999-03-01		Applicant
WO2012/044367	2012-04-01		Applicant

Non-Patent Literature (21)

U.S. Appl. No. 12/565,576, filed Sep. 23, 2009, Kottomtharayil et al.Applicant
U.S. Appl. No. 12/626,839, filed Nov. 27, 2009, Klose.Applicant
Armstead et al., “Implementation of a Campus-Wide Distributed Mass Storage Service: The Dream vs. Reality,” IEEE, 1995, pp. 190-199.Applicant
Arneson, “Mass Storage Archiving in Network Environments” IEEE, 1998, pp. 45-50.Applicant
Cabrera, et al. “ADSM: A Multi-Platform, Scalable, Back-up and Archive Mass Storage System,” Digest of Papers, Compcon '95, Proceedings of the 40th IEEE Computer Society International Conference, Mar. 5, 1995-Mar. 9, 1995, pp. 420-427, San Francisco, CA.Applicant
CommVault Systems, Inc. “Continuous Data Replicator 7.0,” Product Data Sheet, 2007.Applicant
CommVault Systems, Inc., “Deduplication—How To,” http://documentation.commvault.com/release—8—0—0/books—online—1/english—US/features/single—instance—how—to.htm, internet accessed on Jan. 26, 2009, 7 pages.Applicant
CommVault Systems, Inc., “Deduplication,” http://documentation.commvault.com/commvault/release—8—0—0/books—online—1/english—US/features/single—instance/single—instance.htm, internet accessed on Jan. 26, 2009, 9 pages.Applicant
Diligent Technologies HyperFactor, http://www.dilligent.com/products:protecTIER-1:HyperFactor-1, Internet accessed on Dec. 5, 2008, 2 pages.Applicant
Eitel, “Backup and Storage Management in Distributed Heterogeneous Environments,” IEEE, 1994, pp. 124-126.Applicant
Gait, “The Optical File Cabinet: A Random-Access File system for Write-Once Optical Disks,” IEEE Computer, vol. 21, No. 6, pp. 11-22 (1988).Applicant
International Search Report and Written Opinion, International Application No. PCT/US2009/58137, Mail Date Dec. 23, 2009, 14 pages.Applicant
Jander, “Launching Storage-Area Net,” Data Communications, US, McGraw Hill, NY, vol. 27, No. 4(Mar. 21, 1998), pp. 64-72.Applicant
Kornblum, Jesse, “Identifying Almost Identical Files Using Context Triggered Piecewise Hashing,” www.sciencedirect.com, Digital Investigation 3S (2006), pp. S91-S97.Applicant
Lortu Software Development, “Kondar Technology-Deduplication,” http://www.lortu.com/en/deduplication.asp, Internet accessed on Dec. 5, 2008, 3 pages.Applicant
Overland Storage, “Data Deduplication,” http://www.overlandstorage.com/topics/data—deduplication.html, Internet accessed on Dec. 5, 2008, 2 pages.Applicant
Quantum Corporation, “Data De-Duplication Background: A Technical White Paper,” May 2008, 13 pages.Applicant
Rosenblum et al., “The Design and Implementation of a Log-Structure File System,” Operating Systems Review SIGOPS, vol. 25, No. 5, New York, US, pp. 1-15 (May 1991).Applicant
International Search Report and Written Opinion, International Application No. PCT/US2011/30814, Cite Date May 23, 2011.Applicant
International Search Report and Written Opinion, International Application No. PCT/US2011/30804, Cite Date May 24, 2011.Applicant
Office Action in U.S. Appl. No. 12/982,086 dated Apr. 19, 2012.Applicant

Abstract