`
`PATENT 6,415,280
`Docket No. 0100157-00244
`
`Filed on behalf of EMC Corporation and VMware, Inc.
`
`By: Peter Dichiara, Reg. No. 38,005
`David L. Cavanaugh, Reg. No. 36,476
`WILMER CUTLER PICKERING HALE AND DORR LLP
`
`peter.dichiara@wilmerhale.com
`daVid.cavanaugh@wilmerhale.com
`Tel.: 617-526-6466
`
`Fax: 617-526-5000
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`EMC CORPORATION and VMWARE, INC.,
`Petitioners
`
`V.
`
`Patent Owner of
`
`US. Patent No. 6,415,280 to Farber et al.
`
`IPR Case No. IPR2013-00083
`
`REPLY DECLARATION OF DOUGLAS W. CLARK, PH.D.
`
`I, Douglas W. Clark, declare as follows:
`
`1.
`
`I am the same Douglas W. Clark who submitted a prior declaration in
`
`this matter, which I understand was filed on December 15, 2012. My
`
`l
`
`EMC/VMware V. PersonalWeb
`
`IPR2013 -00083
`
`EMCVMW 1 07 8
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`qualifications remain as stated in paragraphs 1-7 and Appendix A of that
`
`declaration, filed as Exhibit 1009 in this case. My statements in paragraphs 8-10
`
`of my prior declaration regarding my review of the ‘280 patent and related
`
`materials also remain unchanged.
`
`2.
`
`Since my prior declaration in this case, I have carefully reviewed
`
`PersonalWeb’s Preliminary Response of March 21, 2013, the Board’s Decision to
`
`Institute of May 17, 2013, the transcript of my deposition taken on July 10 and July
`
`11, 2013, PersonalWeb’s Response of July 24, 2013, the Declaration of Robert
`
`B.K. Dewar of July 24, 2013, and the transcript of the deposition of Robert B.K.
`
`Dewar taken on September 25 and September 26, 2013.
`
`3.
`
`I confirm that everything included in my prior declaration of
`
`December 15, 2012, and all of my testimony given during my deposition of July 10
`
`and July 11, 2013 remain true to the best of my knowledge.
`
`Woodhill discloses “data files” that are “named data items”
`
`4.
`
`I understand that the Board construed a “data file” as a “named data
`
`item, such as a simple file that includes a single, fixed sequence of data bytes or a
`
`compound file that includes multiple, fixed sequences of data bytes.” (Decision at
`
`11.) Woodhill divides files into one or data streams (Woodhill at 4: 13-21; Ex.
`
`1005) and then divides those data streams into one or more binary objects.
`
`(Woodhill at 4:21-23; Ex. 1005.) In other words, Woodhill stores “data files
`
`2
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`comprised of one or more binary objects.” (Woodhill at 2:3 (emphasis added); Ex.
`
`1005.) Accordingly, Woodhill describes “data files” that comprise one binary
`
`object (i.e., a single, fixed sequence of data bytes) and “data files” that comprise
`
`multiple binary objects (i.e., multiple, fixed sequences of data bytes.) For each
`
`binary object, Woodhill calculates a Binary Object Identifier that is “based on the
`
`contents of the binary object so that the Binary Object Identifier 74 changes when
`
`the contents of the binary object changes.” (Woodhill at 8:58-62; Ex. 1005.) For
`
`files comprising a single binary object, the Binary Object Identifier 74 for that
`
`binary object includes a hash of the contents of the entire file.
`
`5.
`
`Dr. Dewar notes that “Woodhill defines a ‘file’ as Viewed by
`
`Distributed Storage Manager program 24 as ‘a collection of data streams,’” and
`
`then concludes that “according to Woodhill’s definition, a ‘file’ that is to be
`
`backed up or subjected to the auditing procedure has at least two data streams.”
`
`(Dewar Decl. at 11 108; Ex. 2013.) I disagree with Dr. Dewar’s conclusion. A
`
`person of ordinary skill in the art would not have read Woodhill to demand that a
`
`file must have at least two data streams. The ordinary meaning of a “collection of
`
`items” is a collection that may include zero, one, or many of those items. For
`
`example, as confirmed by the ‘280 patent, a file system directory is a collection of
`
`files. (‘280 patent at 5:46; Ex. 1001 .) A person of ordinary skill in the art would
`
`understand that a directory may contain zero, one, or many files. Likewise,
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`Woodhill’s statement that a file can be Viewed as “a collection of data streams”
`
`(Woodhill at 4: 13-15; Ex. 1005) is properly interpreted as meaning that a file has a
`
`collection of zero, one or more data streams. Dr. Dewar’s reliance on the plural
`
`form of “data streams” is misplaced given the ordinary meaning of “collection.”
`
`6.
`
`Describing a situation where a file has more than one data stream,
`
`Woodhill discloses that “a file may contain its normal data and may also contain
`
`extended attribute data.” (Woodhill at 4: 18-19; Ex. 1005.) In this case, such a file
`
`would have two data streams: one for its “normal data” and one for its “extended
`
`attribute data.” However, Woodhill is also clear that its system is “for storing data
`
`files comprised of one or more binary objects.” (Woodhill at 2:2-3 (emphasis
`
`added); Ex. 1005.) In the case of a file haVing only a single binary object, one of
`
`ordinary skill in the art would understand that there are four possible cases for this
`
`file: (1) the file does not have “normal data” (e.g., it is an empty file), (2) the file
`
`does not have “extended attribute data,” (3) the file’s “normal data” and “extended
`
`attribute data” are combined in the file’s single binary object, or (4) the file’s
`
`“extended attribute data” is dealt with in some independent way. In each of these
`
`cases, the binary object for that file stores all of the file’s data. Thus, I disagree
`
`with Dr. Dewar’s conclusion that “[e]ven if a ‘file’ in Woodhill were to include
`
`only one ‘binary object’, this does not necessarily mean that the binary object
`
`makes up the entire file.” (Dewar Decl. at 11 107; Ex. 2013.)
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`7.
`
`Furthermore, I point out that Woodhill discloses a File Identification
`
`Record 34 that is saved for each data file that is backed up. (Woodhill at 3 154-57;
`
`Ex. 1005.) Each such record includes a File Name 40 field, which is the “name of
`
`the file.” (Woodhill at 5:61; EX. 1005.) Thus, each of Woodhill’s data files is a
`
`“named data item.” While Dr. Dewar confirms that “File names 40 in Woodhill
`
`identify files,” he goes on to state that Woodhill’s File Name 40 fields do not
`
`identify binary objects (Dewar Decl. at 11 105; EX. 2013).
`
`I disagree that
`
`Woodhill’s File Names 40 do not identify binary objects for cases of files having
`
`only a single binary object. As described above, Woodhill discloses files having
`
`only one binary object. Therefore, for files that have only one binary object, a
`
`person of ordinary skill in the art would understand that the File Name 40 field
`
`names the single binary object for that file. Woodhill fiirther discloses the case
`
`where a “calculated binary object identifier [is] saved as the name of the associated
`
`binary object” (Woodhill at 22:3-4; Ex. 1005), and so binary objects are filrther
`
`named by their Binary Object Identifiers.
`
`Woodhill reguests “data files” using Binary Object Identifiers
`
`8.
`
`I describe below two examples of client requests described by
`
`Woodhill for restoring data files: (1) self-audit requests for a randomly selected
`
`binary object and (2) update requests to restore a granularized file to a previous
`
`version of that file. As described in more detail below, a self-audit request is a
`
`5
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`request for a randomly-selected binary object, the request including the Binary
`
`Object Identifier for that binary object. For files having only a single binary
`
`object, this request restores the entire contents of the file in response to a request
`
`that includes a hash of the entire contents of the file. Also, as described in more
`
`detail below, an update request is a request to restore a current version of a file to a
`
`prior version of that file. In the case where the current version of a file has
`
`significantly changed from the previous version of that file, the entire contents of
`
`the file may be transmitted to the requesting local computer. This scenario may
`
`occur if, for example, a disk failure or virus on a local computer destroys or
`
`overwrites each binary object of a particular file.
`
`9.
`
`A person of ordinary skill in the art would understand, however, that
`
`these are two specific implementations of generally well-known restore techniques.
`
`Indeed, Woodhill notes that “[b]ackup/restore systems have a long history on all
`
`types of computer systems” (Woodhill at 1:23-24; Ex. 1005) and describes its
`
`invention as being “for the management of the storage space on a computer system
`
`[that] provide[s] a backup/restore system.” (Woodhill at 2:40-41; Ex. 1005.)
`
`Accordingly, while Woodhill explicitly describes the implementation of its self-
`
`audit request mechanism (See Woodhill at 18: 10-38; Ex. 1005), the purpose of this
`
`mechanism is “to ensure that the binary objects that have been backed up can be
`
`restored.” (Woodhill at 18:12-13; Ex. 1005 .) In other words, the self-audit request
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`mechanism ensures, before a disk failure or a disaster strikes, that it will be
`
`possible to restore files after such an event. Indeed, the very goal of Woodhill’s
`
`backup procedure is to “ensure[] that at least one copy of every binary object is
`
`stored and that a disaster that destroys an entire site would not destroy all copies of
`
`that site's data.” (Woodhill at 9:42-44; EX. 1005.) Thus, it is clear that Woodhill
`
`includes the ability to restore files previously backed up. Rather than explain in
`
`detail the general case of such a well-known procedure, Woodhill describes the
`
`implementation of its more complicated “update request” procedure for restoring
`
`granularized binary objects. (See Woodhill at 17: 18-1819; EX. 1005.) Because (as
`
`described below) an “update request” includes Binary Object Identifiers for the
`
`binary objects of a file to be restored and current granule contents identifiers
`
`(Woodhill at 17:42-46; Ex. 1005), it is clear that the more general case of restoring
`
`a non-granularized file would include Binary Object Identifiers for the binary
`
`objects of that file. The only difference between these cases is that granule
`
`contents identifiers would not be necessary for non-granularized file. One skilled
`
`in the art would appreciate that a “back-up” writes a version of a file to the back-up
`
`server, and that a “restore” reads a version of a file from the back-up server. These
`
`are well-understood data management techniques that existed long before the
`
`patents.
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`10. Despite the existence of well-known mechanisms for restoring data
`
`files, Dr. Dewar argues that “Woodhill fails to disclose a client request for a data
`
`file, where the ‘request’ includes ‘a hash of the contents of the data file.
`
`(Dewar
`
`’7)
`
`Decl. at 11 102; Ex. 2013.) In particular, Dr. Dewar argues that, “in Woodhill’s
`
`self-auditing procedure there is no description of a ‘request’ for a binary object that
`
`includes [Binary Object Identification] Record 5 8.” (Dewar Decl. at 11 104; Ex.
`
`2013.) However, as I described in my prior declaration for this case, Woodhill
`
`indeed requests binary objects using the Binary Object Identifier 74 portion of the
`
`corresponding Binary Object Identification Record 58 during its self-audit
`
`procedure. For example, during Woodhill’s self-audit procedure, the “Distributed
`
`Storage Manager program 24 initiates a restore of a randomly selected binary
`
`object identified by a Binary Object Identification Record 58 stored in File
`
`Database 25.” (Woodhill at 18:17-19; Ex. 1005.) A person of ordinary skill in the
`
`art would understand this sentence to mean that Woodhill randomly selects a
`
`Binary Object Identification Record 58 from the potentially many such records
`
`stored in File Database 25, and requests the binary object corresponding to that
`
`record. For such a routine operation, there was no need for Woodhill to explicitly
`
`describe that the binary object identifier from the record is included as part of a
`
`request. Rather, it was well-known at the time of the Woodhill patent how to
`
`request objects using their identifiers. In my prior declaration for this case, I
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`mentioned many such examples, including the Langer reference (which dates back
`
`to 1991). PersonalWeb seems to have conceded this point by noting that “Langer
`
`appears to disclose accessing a standalone file by employing an MD5 of the file
`
`contents.” (IPR2013-00085, Resp. 41; Ex. 1079.) Accordingly, a person of
`
`ordinary skill in the art would understand that Woodhill’s remote backup file
`
`server can provide binary objects in response to their identifiers. Woodhill’s
`
`Figure 3 makes clear that the Binary Object Identifier 74 portion of the Binary
`
`Object Identification Record 58 is the key part of that record. (Woodhill at Fig. 3;
`
`Ex. 1005.)
`
`11. During Woodhill’s self-audit procedure, the binary object is
`
`“identified by a Binary Object Identification Record 58.” (Woodhill at 18: 18-19;
`
`Ex. 1005.) As described above, a person of ordinary skill in the art would
`
`understand that the key component of this record is the Binary Object Identifier 74.
`
`(Woodhill at Fig. 3; Ex. 1005.) However, rather than acknowledge the role of
`
`Binary Object Identifier 74, Dr. Dewar lays out a sequence of steps using the other
`
`components of the Binary Object Identification Record 58, and argues that “[t]he
`
`‘Link to Backup Instance Record 60’ field is the field used in Woodhill's self-
`
`auditing procedure to reach the ‘Link to File Identification Record 44.’ And the
`
`‘Link to File Identification Record 44’ is then used to reach ‘File Identification
`
`Record 34.’ The ‘File Name 40’ and ‘File Location 38’ fields in the ‘File
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`Identification Record 34’ are used to access the file containing the binary object at
`
`backup storage. Then, the "Binary Object Stream Type 62" and "Binary Object
`
`Offset 72" are used to locate the binary object in that file.” (Dewar Decl. at 11 5 3;
`
`Ex. 2013.) Because Woodhill’s self-audit procedure is capable of restoring a
`
`binary object stored on the remote backup file server, it would not make sense to
`
`use this sequence of steps to access a binary object. First, the Binary Object
`
`Stream Type field 62 and the Binary Object Offset field 72 would not be useful in
`
`locating a binary object, as these fields do not identify any particular binary object.
`
`Second, the Link to Backup Instance Record 60 field links to another record stored
`
`in the File Database, which is a database stored and maintained on a local
`
`computer. (Woodhill at 3:45-47; EX. 1005.) Thus, the linked record would not be
`
`available on the remote backup file server, unless the corresponding Backup
`
`Instance Record was also sent from the local computer to the remote backup file
`
`server (which would not be efficient, and which Woodhill does not suggest in any
`
`event). Furthermore, Dr. Dewar suggests using not just one link (the Link to
`
`Backup Instance Record 60), but a second link (the Link to File Identification
`
`Record 44) to access yet another record. (Dewar Decl. at 11 53; Ex. 2013.) Again,
`
`these records are all stored on a local computer’s File Database, and so it would
`
`not make sense that they would be used to access a binary object from the remote
`
`backup file server in this way. Even if these fields were accessible by, or
`
`10
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`transmitted to, the remote backup file server, the File Name 40 and File Location
`
`38 fields represent the local location of the file on a local computer, not on the
`
`remote backup file server. Thus, they would not be usefill in accessing a binary
`
`object on the remote backup file server. Put more simply, the fields of a Binary
`
`Object Identification Record 58, other than the Binary Object Identifier 74 portion,
`
`are simply not relevant to accessing a binary object on the remote backup file
`
`server.
`
`12.
`
`I also point out that Woodhill’s self-audit procedure is clearly initiated
`
`by a local computer. Woodhill describes that, “the Distributed Storage Manager
`
`program 24 initiates a restore of a randomly selected binary object identified by a
`
`Binary Object Identification Record 58 stored in File Database 25.” (Woodhill at
`
`18:16-19 (emphasis added); Ex. 1005.) Woodhill is clear that its “discussion
`
`illustrates the operation of the Distributed Storage Manager program 24 on a single
`
`local computer 20” and that “it should be understood that the Distributed Storage
`
`Manager program 24 operates in the same fashion on each local computer 20.”
`
`(Woodhill at 5:4-8; Ex. 1005.) Therefore, one of ordinary skill in the art would
`
`understand that the self-audit procedure is clearly initiated by the Distributed
`
`Storage Manager program 24 as it executes on a local computer. This is fiirther
`
`confirmed by Woodhill’s reference to initiating the self-audit using “a Binary
`
`Object Identification Record 58 stored in File Database 25.” (Woodhill at 18:18-
`
`11
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`19 (emphasis added); EX. 1005.) Woodhill is clear that “[t]he Distributed Storage
`
`Manager program 24 of the present invention builds and maintains the File
`
`Database 25 on one of the disk drives 19 on each local computer 20” (Woodhill at
`
`3:45 -47 ; Ex. 1005), and so the self-audit procedure initiates the self-audit using the
`
`Binary Object Identification Record 58 from the File Database 25 on the local
`
`computer. As described above, the Binary Object Identifier 74 portion of this
`
`record is then transmitted to the remote backup file server. Each Binary Object
`
`Identifier 74 includes a Binary Object Hash 70 field. (Woodhill at Fig. 3; Ex.
`
`1005.) Accordingly, when a local computer initiates the self-audit procedure, a
`
`binary object is requested using a hash of the contents of the binary object. In the
`
`case of a data file having a single binary object, a self-audit request for that binary
`
`object is a request for that data file. A self-audit request includes a hash of the
`
`contents of the binary object, which, in this case, is a hash of the contents of the
`
`entire data file.
`
`13.
`
`Because Woodhill requests binary objects using their Binary Object
`
`Identifiers, I also disagree with Dr. Dewar’s conclusion that Woodhill’s Binary
`
`Object Identifiers are “simply used for comparison purposes after the binary object
`
`has already been accessed.” (Dewar Decl. at 11 103; Ex. 2013.) Woodhill’s Binary
`
`Object Identifiers are called “identifiers” precisely because they are used for more
`
`than comparison. As described above, Binary Object Identifiers are used for
`
`12
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`requesting binary objects and Woodhill’s claim 1 even makes clear that a binary
`
`object identifier is “saved as the name of the associated binary object.” (Woodhill
`
`at 22:3-4; Ex. 1005.)
`
`14.
`
`Furthermore, in the more general case of a single binary object being
`
`requested where that binary object belongs to a file comprising many binary
`
`objects (i.e., a compound file), Woodhill works in the same way as the ‘280 patent.
`
`For a read operation involving a compound file, the ‘280 patent’s Read File
`
`mechanism “break[s] the read operation into one or more read operations on
`
`component segments” by “[i]dentify[ing] the segment(s) to be read” and “[u]s[ing]
`
`the Read File mechanism (recursively) to read from [each] segment.” (‘280 patent
`
`at 21 :30-50; Ex. 1001.) Similarly, the ‘280 patent makes clear that its “Request
`
`True File remote mechanism copies only a single data item from one processor to
`
`another. If the data item is a compound file, its component segments are not
`
`copied, only the indirect block is copied. The segments are copied only when they
`
`are read (or otherwise needed).” (‘280 patent at 34:4-8; Ex. 1001.) In both of
`
`these cases, requests for compound files do not cause the underlying file data to be
`
`retrieved. Instead, requests including compound file hashes merely retrieve an
`
`indirect block of segment hashes. Segment data is only provided in response to
`
`subsequent requests that include those segment hashes. Like the ‘280 patent’s
`
`requests for data of a single segment of a compound file, Woodhill’s self-audit
`
`l3
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`request may retrieve a single binary object fiom a file comprising multiple binary
`
`objects.
`
`15. As another example of requesting a data file using Binary Object
`
`Identifiers 74, Woodhill discloses an “update request” that is transmitted to the
`
`remote backup file server. (Woodhill at 17:41-43; Ex. 1005.) The update request
`
`is used to “update” a file as it currently exists on a local computer to a version that
`
`was previously backed up. (See Woodhill at 17:18-43; Ex 1005.) The update
`
`request works for files, such as large database files, that were previously backed up
`
`using Woodhill’s “granularization” procedure. (Woodhill at 14:53-65; Ex. 1005 .)
`
`Woodhill is explicit that each update request “includes the Binary Object
`
`Identification Record 58 for the previous version of each binary object.”
`
`(Woodhill at 17:41-45; Ex. 1005.) Each of these Binary Object Identification
`
`Records 58 includes a Binary Object Identifier 74 portion and each Binary Object
`
`Identifier 74 includes a Binary Object Hash 70 value. (Woodhill at Fig. 3; Ex.
`
`1005.) Therefore, an update request includes a hash of the contents of the file
`
`being requested. Also included in an “update request” are “‘contents identifiers’
`
`for each ‘granule’ within the current version of each binary object as it exists on
`
`the local computer 20.” (Woodhill at 17:41-47; Ex. 1005.) Woodhill’s remote
`
`backup file server proceeds to compare the granule contents identifiers for the
`
`current version of each binary object with granule contents identifiers
`
`l4
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`corresponding to the requested previous version of those binary objects. (Woodhill
`
`at 17:50-55; Ex. 1005.) In each case, when the contents identifiers “do not match, .
`
`.
`
`. the Distributed Storage Manager program 24 transmits the ‘granule’ to the local
`
`computer 20.
`
`After all ‘granules’ are received from the remote backup file
`
`server 12, the binary object has been restored to the state of the previous version.”
`
`(Woodhill at 17:62-18:9; Ex. 1005.) Therefore, for any of the requested previous
`
`versions of the binary objects, each granule that differs fiom the version as it
`
`existed on the local computer (at the time of the request) is sent to that local
`
`computer. Accordingly, if every granule of a previous version of a binary object
`
`differed from the corresponding current granule, every granule of that binary
`
`object would be transmitted from the remote backup file server to the requesting
`
`local computer. In this way, the full contents of a previous version of a binary
`
`object would be restored in response to a request including the Binary Object
`
`Identifier 74 of that binary object. Therefore, the filll contents of a previous
`
`version of a file may be transmitted from the remote backup file server to the
`
`requesting local computer in this way (e.g., if every granule of every binary object
`
`for that file differed fiom the corresponding granule that was previously backed
`
`up). Even if some granules are the same, the full contents of a file are nonetheless
`
`restored by the update request mechanism.
`
`15
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`Woodhill backs up files regardless of their number of binary objects
`
`16.
`
`Dr. Dewar argues that “Woodhill’s definition of ‘file’ ...requires that
`
`each ‘file’ to be backed up includes at least two binary objects.” (Dewar Decl. at 11
`
`108; Ex. 2013.) However, explained above, Woodhill clearly discloses files
`
`having only a single binary object. Furthermore, Woodhill in no way suggests that
`
`files with only a single binary object should be treated any differently than files
`
`having two or more binary objects. In fact, if Woodhill implemented an arbitrary
`
`distinction that prevented files with only a single binary object from being backed
`
`up or audited, system performance could dramatically suffer. For example, a
`
`document may be very small (and thus fit within a single binary object), yet
`
`nonetheless be very important. For example, a resume or a copy of a lease may be
`
`only a few pages long (and thus likely fit well within a single binary object), while
`
`still being one of the most important documents stored on a user’s computer. A
`
`user would expect that backup copies of these important documents (regardless of
`
`their number of binary objects) would be stored so that they could be safely
`
`recovered if a disaster destroyed the user’s computer or damaged its disk drives.
`
`Woodhill’s system is designed to “ensure[] that at least one copy of every binary
`
`object is stored and that a disaster that destroys an entire site would not destroy all
`
`copies of that site’s data.” (Woodhill at 9:42-44; Ex. 1005.) Likewise, the self-
`
`audit procedure “ensure[s] that the binary objects that have been backed up can be
`
`16
`
`
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157—00244
`
`restored.” (Woodhill at 18: 12-13; Ex. 1005.) If Woodhill only backed up or self-
`
`audited files having two or more binary objects (as Dr. Dewar has claimed), then
`
`Woodhill’s system would not effectively prevent a disaster from destroying all
`
`copies of those files having only a single binary object. Therefore, a person of
`
`ordinary skill in the art would not read Woodhill to demand that files to be backed
`
`up or subject to the auditing procedure are in any way distinct from files otherwise
`
`described by Woodhill.
`
`Availabilifl for Cross-Examination
`
`17.
`
`In signing this declaration, I recognize that the declaration will be
`
`filed as evidence in a contested case before the Patent Trial and Appeal Board of
`
`the United States Patent and Trademark Office. I also recognize that I may be
`
`subject to cross examination in the case and that cross examination will take place
`
`within the United States. If cross examination is required of me, I will appear for
`
`cross examination within the United States during the time allotted for cross
`
`examination.
`
`Right to Supplement
`
`18.
`
`I reserve the right to supplement my opinions in the fiiture to respond
`
`to any arguments that Patentee raises and to take into account new information as it
`
`becomes available to me.
`
`17
`
`
`
`J urat
`
`IPR2013-00083
`
`PATENT 6,415,280
`Docket No. 0100157-00244
`
`19.
`
`I declare that all statements made herein of my own knowledge are
`
`true and that all statements made on information and belief are believed to be true;
`
`and further that these statements were made with the knowledge that willful false
`
`statements and the like so made are punishable by fine or imprisonment, or both,
`
`under Section 1001 of Title 18 of the United States Code.
`
`Dated: October 1,2013
`
`Z
`
`Z Z
`
`Douglas W. Clark
`Philadelphia, PA
`
`18
`
`