`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`Hewlett Packard Enterprise Co.,
`Petitioner,
`v.
`Intellectual Ventures I LLC,
`Patent Owner.
`
`
`Patent No. 6,779,082 to Burger et al.
`Issued: August 17, 2004
`Filed: February 5, 2001
`Title: Network-Based Disk Redundancy Storage System and Method
`IPR Case No.: IPR2022-00290
`
`
`DECLARATION OF TODD MOWRY, PH.D.
`
`1
`
`
`
`HPE, Exh. 1003, p. 1
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`TABLE OF CONTENTS
`TABLE OF EXHIBITS ............................................................................................. 4
`I.
`INTRODUCTION ........................................................................................... 6
`II.
`QUALIFICATIONS ........................................................................................ 7
`III. LEGAL STANDARDS ................................................................................. 11
`A. Anticipation ......................................................................................... 11
`B.
`Obviousness ......................................................................................... 12
`IV. LEVEL OF ORDINARY SKILL IN THE ART ........................................... 15
`V. KNOWLEDGE OF A POSITA ..................................................................... 17
`A. Known Distributed Storage Systems .................................................. 17
`B.
`Redundancy and Selecting Storage Devices for Client Access .......... 21
`VI. OVERVIEW OF THE ’082 PATENT .......................................................... 23
`A.
`Priority Date ........................................................................................ 23
`B.
`The ’082 Patent Specification ............................................................. 24
`C.
`The Challenged Claims ....................................................................... 27
`1.
`Claim 1 ..................................................................................... 28
`2.
`Claim 2 ..................................................................................... 29
`3.
`Claim 3 ..................................................................................... 30
`4.
`Claim 4 ..................................................................................... 30
`5.
`Claim 5 ..................................................................................... 30
`6.
`Claim 9 ..................................................................................... 30
`7.
`Claim 10 ................................................................................... 31
`8.
`Claim 11 ................................................................................... 31
`9.
`Claim 12 ................................................................................... 32
`The ’082 Patent Prosecution History .................................................. 32
`D.
`VII. CLAIM CONSTRUCTION .......................................................................... 34
`VIII. OVERVIEW OF PRIOR ART ...................................................................... 35
`A. Moore (Ex. 1005) ................................................................................ 35
`
`
`
`2
`
`
`
`HPE, Exh. 1003, p. 2
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`B. Moore Publication (Ex. 1006) ............................................................. 41
`C.
`Coates (Ex. 1004) ................................................................................ 42
`IX. ANALYSIS .................................................................................................... 49
`A. GROUNDS 1-3: Claims 1-5 and 9-12 Are Anticipated by Moore,
`Rendered Obvious by Moore in View of the Knowledge of a POSITA,
`or Rendered Obvious by Moore in View of the Moore Publication and
`the Knowledge of a POSITA .............................................................. 50
`1.
`Challenged Claims: .................................................................. 50
`GROUNDS 4 AND 5: Claims 1, 5 and 9 Are Anticipated by Coates
`and Claims 1-5 and 9-12 Are Rendered Obvious by Coates in View of
`the Knowledge of a POSITA .............................................................. 75
`1.
`Challenged Claims ................................................................... 76
`X. DECLARATION IN LIEU OF OATH .......................................................106
`
`
`
`B.
`
`
`
`3
`
`
`
`HPE, Exh. 1003, p. 3
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`TABLE OF EXHIBITS
`
`Exhibit
`
`Description
`
`1001
`
`1002
`
`1003
`
`1004
`
`1005
`
`1006
`
`1007
`
`1008
`
`1009
`
`1010
`
`1011
`
`U.S. Patent No. 6,779,082 (“the ’082 Patent”)
`
`File History of the ’082 Patent
`
`Declaration of Todd Mowry, Ph.D
`
`U.S. Patent No. 7,266,556 to Coates (“Coates”)
`
`U.S. Patent No. 6,678,700 to Moore et al. (“Moore”)
`
`Moore, Reagan, Chaitan Baru, Amarnath Gupta, Bertram
`Ludaescher, Richard Marciano, and Arcot Rajasekar. Collection-
`Based Long-Term Preservation. San Diego, CA: San Diego
`Supercomputer Center, 1999. (“Moore Publication”)
`
`Hartman, John H., and John K. Ousterhout. “The Zebra Striped
`Network File System.” ACM Transactions on Computer Systems,
`vol. 3, issue 3 (1 August 1995): 274-310.
`
`Floyd, Richard Allen. Transparency in Distributed File Systems.
`Technical Report 272, January 1989. Rochester, NY: University of
`Rochester, Dept. of Computer Science, 1988.
`
`U.S. Patent No. 6,185,598 to Farber et al.
`
`Patterson, David A., Garth Gibson, and Randy H. Katz. “A Case
`for Redundant Arrays of Inexpensive Disks (RAID).” In
`Proceedings of the 1988 ACM SIGMOD International Conference
`on Management of Data (pp. 109-116). Berkeley, CA: Computer
`Science Division, University of California, [1987]
`
`Gibson, Garth A., David F. Nagle, Khalil Amiri, Fay W. Chang,
`Eugene M. Feinberg, Howard Gobioff, Chen Lee, Berend Ozceri,
`Erik Riedel, David Rochberg, and Jim Zelenka. “File Server
`Scaling with Network-Attached
`Secure Disks.”
`In
`SIGMETRICS97:
`ACM
`SIGMETRICS Conference
`on
`
`
`
`4
`
`
`
`HPE, Exh. 1003, p. 4
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`Exhibit
`
`1012
`
`1013
`
`1014
`
`1015
`
`1016
`
`Description
`Measurement & Modeling of Computer Systems (pp. 272-284).
`Sigmetrics ‘97. New York: Association for Computing Machinery,
`1997.
`
`Bolosky, William J., John R. Douceur, David Ely, and Marvin
`Theimer. “Feasibility of a Serverless Distributed File System
`Deployed on an Existing Set of Desktop PCs.” ACM SIGMETRICS
`Performance Evaluation Review, vol. 28, no. 1 (June 2000): 34–
`43.
`
`Declaration of Sylvia D. Hall-Ellis, Ph.D.
`
`the Random House Webster’s Unabridged
`Excerpts from
`Dictionary, Second Edition (2001)
`
`Excerpts from The American Heritage College Dictionary, Third
`Edition (2000)
`
`Lex Machina data for trials scheduled in the Western District of
`Texas
`
`
`
`5
`
`
`
`HPE, Exh. 1003, p. 5
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`I, Todd Mowry, Ph.D., declare as follows:
`
`I.
`
`INTRODUCTION
`1.
`I am over the age of 21 and am competent to make this declaration.
`
`2.
`
`I have been retained on behalf of Hewlett Packard Enterprise Co.
`
`(“HPE”) to provide my opinions regarding the validity of claims 1-5 and 9-12 of
`
`U.S. Patent No. 6,779,082 (“’082 Patent”). I submit this declaration based on my
`
`personal knowledge and in support of HPE’s inter partes review petition against the
`
`’082 Patent (the “Petition”).
`
`3.
`
`I have been asked to review and provide my independent analysis of
`
`the ’082 Patent in light of the materials cited below and my knowledge and
`
`experience in this field during the relevant period. I have been asked to consider
`
`whether the references cited in the Petition anticipate and/or render obvious claims
`
`1-5 and 9-12 of the ’082 Patent.
`
`4.
`
`I am being compensated according to my normal hourly rate for my
`
`time providing my independent analysis in this aforementioned IPR proceeding, but
`
`my compensation is not contingent in any way on the content of my analysis or the
`
`outcome of this proceeding. I am not, and never was, an employee or agent of HPE.
`
`5. My findings, as explained below, are based on my study, experience,
`
`and background discussed below, informed by my extensive experience in the field
`
`
`
`6
`
`
`
`HPE, Exh. 1003, p. 6
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`of storage systems, including distributed and redundant storage systems. My
`
`findings are also based on my education as a computer scientist and electrical
`
`engineer, in addition to the subsequent decades of work in research and development
`
`in these fields. As described in more detail below, based on my experiences, I
`
`understand and know of the capabilities of persons of ordinary skill in the field of
`
`storage systems, including distributed and redundant storage systems, in 2001, when
`
`the application to which the ’082 Patent claims priority was filed. Indeed, I have
`
`extensive relevant personal knowledge and experience, in addition to working
`
`directly with many such persons in these fields during that time frame. I have also
`
`relied on my review and analysis of the prior art cited in the Petition and the other
`
`materials cited herein, including those itemized in the “Table of Exhibits” list
`
`preceding this declaration.
`
`II. QUALIFICATIONS
`6.
`Attached as Appendix A is my curriculum vitae, which includes a more
`
`detailed statement of my professional qualifications,
`
`including education,
`
`publications, honors and awards, professional activities, consulting engagements,
`
`and other relevant experience. While I incorporate Appendix A by reference, below
`
`is a brief summary of my background, including my background and experience
`
`relevant to this case.
`
`
`
`7
`
`
`
`HPE, Exh. 1003, p. 7
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`7.
`I am a Professor in the Department of Computer Science at Carnegie
`
`Mellon University. I also have a courtesy appointment in the Department of
`
`Electrical and Computer Engineering. I have served on the faculty of Carnegie
`
`Mellon University for 24 years starting in 1997 through the present (2021).
`
`8.
`
`I also served on the faculty of the University of Toronto for four years
`
`between 1993 and 1997, in the Department of Electrical and Computer Engineering
`
`and a courtesy appointment in the Department of Computer Science. Prior to that
`
`appointment, I was a Graduate Research Assistant in the Department of Electrical
`
`Engineering at Stanford University for four years between 1989 and 1993.
`
`9.
`
`As a faculty member, I have taught and continue to teach courses and
`
`directed research in computer systems and software, operating systems, storage
`
`systems and databases, distributed systems, parallel processing, compiler
`
`optimization, and computer architecture.
`
`10.
`
`I received a B.S. degree in Electrical Engineering with Highest
`
`Distinction from the University of Virginia in May 1988. I received an M.S. in
`
`Electrical Engineering from Stanford University in June 1989, and a Ph.D. in
`
`Electrical Engineering from Stanford University in March 1994.
`
`11.
`
`I have worked in the computer industry in various capacities. I was a
`
`part-time Computer Architect and then Computer Architecture Consultant at Silicon
`
`
`
`8
`
`
`
`HPE, Exh. 1003, p. 8
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`Graphics, Inc., in Mountain View, California (formerly MIPS Computer Systems in
`
`Sunnyvale, California) from 1989 to 1993 and 1993 to 1996, respectively. I was a
`
`Visiting Scientist at IBM in Toronto from 1996 to 2004. During that same time
`
`period (1996 to 2004), I was also a Member of the Technical Advisory Board of
`
`SandCraft, Inc. in Santa Clara, California. I was the Director of the Intel Research
`
`Pittsburgh Lab at Intel Corporation in Pittsburgh, Pennsylvania from 2004 to 2007.
`
`12.
`
`I am a member of the following research groups at Carnegie Mellon
`
`University: the Parallel Data Laboratory (academia’s premier storage systems
`
`research center, which focuses on storage systems, databases, cloud computing,
`
`etc.), the Carnegie Mellon Database Group, and the Catalyst group (which combines
`
`research in machine learning and systems to accelerate AI applications).
`
`13.
`
`I have authored 19 journal articles and 60 conference papers. I am also
`
`an inventor on 5 patents.
`
`14.
`
`I am the recipient of several honors and awards: the Arthur Samuel
`
`Thesis Award (awarded by the Stanford Computer Science department to the top
`
`two Ph.D. theses in a given year), several IBM Faculty Development Awards (1996,
`
`1997, 1998, 2000, 2001, 2002, and 2003), several Best Paper Awards (the Second
`
`Symposium on Operating Systems Design and Implementation in 1996; the 20th
`
`International Conference on Data Engineering (ICDE) in 2004), the Alfred P. Sloan
`
`
`
`9
`
`
`
`HPE, Exh. 1003, p. 9
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`Research Fellowship (awarded to researchers in recognition of distinguished
`
`performance and a unique potential to make substantial contributions to their field),
`
`the Most Thought-Provoking Idea Award in 2004 (awarded by the Architectural
`
`Support for Programming Languages and Operating Systems (ASPLOS)), the
`
`TR100 Award in 1999 (awarded by MIT’s Technology Review magazine to the top
`
`100 most promising young innovators in science and technology), and I became an
`
`ACM Fellow in 2016.
`
`15.
`
`I am a member of the Institute of Electrical and Electronics Engineers
`
`(IEEE) and the Association of Computing Machinery (ACM). I was the Editor-in-
`
`Chief of ACM Transactions on Computer Systems, the premier journal for computer
`
`systems research, from 2013-2018. I was an Associate Editor for the journal prior
`
`to that, since 2001. I was the Program Chair of the International Conference on
`
`Architectural Support for Programming Languages and Operating Systems
`
`(ASPLOS) in 2010. I was the Co-Program Chair of the International Conference on
`
`Parallel Architectures and Compilation Techniques (PACT) in 2001. I have been on
`
`the programming committee in various years for ASPLOS, the International
`
`Symposium on Computer Architecture (ISCA), and the International Symposium on
`
`Microarchitectures.
`
`
`
`10
`
`
`
`HPE, Exh. 1003, p. 10
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`16. Overall, I have over 25 years of experience in the field of computer
`
`science.
`
`III. LEGAL STANDARDS
`17.
`I understand that the prior art references cited herein qualify as prior art
`
`to the ’082 Patent under U.S. law. In formulating my opinions, I have applied certain
`
`legal standards.
`
`18.
`
`I am not a lawyer. I do not expect to offer any testimony regarding
`
`what the law is. Instead, the following sections summarize the law as I understand
`
`it in formulating and rendering my opinions found in this declaration. I understand
`
`that, in an inter partes review proceeding, patent claims may be deemed
`
`unpatentable if it is shown that they were anticipated by a single patent or printed
`
`publication or rendered obvious by one or more prior art patents or printed
`
`publications. I understand that questions of claim clarity (definiteness) and
`
`enablement cannot be considered as a ground for considering the patentability of a
`
`claim in these proceedings.
`
`A. Anticipation
`19.
`I understand that a patent claim may be invalid as “anticipated” if each
`
`and every feature of the claim is found, expressly or inherently, in a single item of
`
`prior art, such as in a single prior art patent or printed publication. In determining
`
`
`
`11
`
`
`
`HPE, Exh. 1003, p. 11
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`whether the single item of prior art anticipates the claim, one considers not only what
`
`is expressly/implicitly disclosed in the particular item of prior art, but also what is
`
`inherently present or disclosed in that prior art or what inherently results from its
`
`practice.
`
`B. Obviousness
`20.
`I understand that a claim may be invalid if the subject matter described
`
`by the claim as a whole would have been obvious to a person of ordinary skill in the
`
`art (“POSITA”) in view of a prior art reference, or in view of a combination of
`
`references at the time the claimed invention was made. Therefore, I understand that
`
`obviousness is determined from the perspective of a POSITA, and that the asserted
`
`claims of the patent should be read from the point of view of such a person at the
`
`time the claimed invention was made. I further understand that a POSITA is
`
`assumed to know and to have all relevant prior art in the field of endeavor covered
`
`by the patent in suit and all analogous prior art. I understand that obviousness in an
`
`inter partes review proceeding is evaluated using a preponderance of the evidence
`
`standard, which means that the claims must be more likely obvious than nonobvious.
`
`21.
`
`I also understand that an analysis of whether a claimed invention would
`
`have been obvious should be considered in light of the scope and content of the prior
`
`art, the differences (if any) between the prior art and the claimed invention, and the
`
`
`
`12
`
`
`
`HPE, Exh. 1003, p. 12
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`level of ordinary skill in the pertinent art involved. I understand as well that a prior
`
`art reference should be viewed as a whole. I understand that in considering whether
`
`an invention for a claimed combination would have been obvious, I may assess
`
`whether there are apparent reasons to combine known elements in the prior art in the
`
`manner claimed in view of interrelated teachings of multiple prior art references, the
`
`effects of demands known to the design community or present in the marketplace,
`
`and/or the background knowledge possessed by a POSITA. I also understand that
`
`other principles may be relied on in evaluating whether a claimed invention would
`
`have been obvious, and that these principles include the following:
`
`
`
`A combination of familiar elements according to known methods is
`
`likely to be obvious when it does no more than yield predictable results;
`
` When a device or technology is available in one field of endeavor,
`
`design incentives and other market forces can prompt variations of it,
`
`either in the same field or in a different one, so that if a POSITA can
`
`implement a predictable variation, the variation is likely obvious;
`
`
`
`If a technique has been used to improve one device, and a POSITA
`
`would have recognized that it would improve similar devices in the
`
`same way, using the technique is obvious unless its actual application
`
`is beyond his or her skill;
`
`
`
`13
`
`
`
`HPE, Exh. 1003, p. 13
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`
`
`An explicit or implicit teaching, suggestion, or motivation to combine
`
`two prior art references to form the claimed combination may
`
`demonstrate obviousness, but proof of obviousness does not depend on
`
`or require showing a teaching, suggestion, or motivation to combine;
`
`Market demand, rather than scientific literature, can drive design trends
`
`and may show obviousness;
`
`One of the ways in which a patent’s subject can be proved obvious is
`
`by noting that there existed at the time of invention a known problem
`
`for which there was an obvious solution encompassed by the patent’s
`
`
`
`
`
`claims;
`
`
`
`Any need or problem known in the field of endeavor at the time of
`
`invention and addressed by the patent can provide a reason for
`
`combining the elements in the manner claimed;
`
`
`
`“Common sense” teaches that familiar items may have obvious uses
`
`beyond their primary purposes, and in many cases a POSITA will be
`
`able to fit the teachings of multiple patents together like pieces of a
`
`puzzle;
`
`
`
`A POSITA is also a person of ordinary creativity, not an automaton;
`
`
`
`14
`
`
`
`HPE, Exh. 1003, p. 14
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`
`
`A patent claim can be proved obvious by showing that the claimed
`
`combination of elements was “obvious to try,” particularly when there
`
`is a design need or market pressure to solve a problem and there are a
`
`finite number of identified, predictable solutions such that a POSITA
`
`would have had good reason to pursue the known options within his or
`
`her technical grasp; and
`
`
`
`One should be cautious of using hindsight in evaluating whether a
`
`claimed invention would have been obvious.
`
`IV. LEVEL OF ORDINARY SKILL IN THE ART
`22.
`I understand that my assessment of the claims of the ’082 Patent and
`
`the teachings of the prior art and my analysis and opinions herein must be undertaken
`
`from the perspective of what would have been known or understood by a person
`
`having ordinary skill in the art, reading the ’082 Patent on its priority date and in
`
`light of the specification and file history of the ’082 Patent. I will refer to such a
`
`person as a “POSITA.”
`
`23.
`
`I further understand that in determining the level of ordinary skill in the
`
`art, I am to consider factors including:
`
`(a) the type of problems encountered in the art or field of invention,
`(b) prior art solutions to those problems,
`
`
`
`15
`
`
`
`HPE, Exh. 1003, p. 15
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`(c) the rapidity with which innovations are made,
`(d) sophistication of the technology, and
`(e) the educational level of active workers in the field.
`24.
`I understand that a person of ordinary skill in the art is not a specific
`
`real individual, but rather a hypothetical individual having the qualities reflected by
`
`the factors above. This hypothetical person has knowledge of all prior art in the
`
`relevant field and takes from each reference what it would teach to a person having
`
`the skills of a POSITA.
`
`25.
`
`I understand that a POSITA is a person of ordinary creativity, but not
`
`an automaton, and that a POSITA can often fit multiple patents or prior art references
`
`together like pieces of a puzzle as a result of this ordinary creativity. I also
`
`understand that I may consider the inferences and creative steps that a POSITA
`
`would employ. In addition, I understand that a POSITA would necessarily have
`
`been capable of understanding the scientific and engineering principles applicable
`
`to the pertinent art.
`
`26. Based on my review and analysis of the ’082 Patent, the prior art cited
`
`herein, and the ordinary skill factors described in this section, a POSITA in the field
`
`of the ’082 Patent at the time of the earliest priority date (February 5, 2001) would
`
`have had at least a bachelor’s degree in electrical engineering, computer science, or
`
`
`
`16
`
`
`
`HPE, Exh. 1003, p. 16
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`an equivalent, plus three years of experience in the field of storage system design
`
`and architecture. Additional education in the fields of electrical engineering or
`
`computer science such as a masters or doctorate degree may serve as a substitute for
`
`experience in the field. A person with less education but more relevant practical
`
`experience may also meet this standard. The prior art also evidences the level of
`
`skill in the art.
`
`27. As of February 5, 2001, i.e., the earliest priority date of the ’082 Patent,
`
`I had more than ordinary skill in the art. I am, however, familiar with the skills and
`
`knowledge possessed by those I would have considered a POSITA as of that date.
`
`When I refer to the understanding of a POSITA, I am referring to the understanding
`
`of such a person as of February 5, 2001.
`
`V. KNOWLEDGE OF A POSITA
`A. Known Distributed Storage Systems
`28. The concept of distributing data storage over devices (such as servers,
`
`disk arrays, computer databases) connected to a network has been known for
`
`decades. E.g., Ex. 1008. As networks expanded in the 1990s, growing amounts of
`
`data moved from local storage—i.e., storage located on the same device as the
`
`requesting process—to networked devices remote from the requesting device. Id. at
`
`19 (addressing file transfer over LANs); Ex. 1004 at 5:11-26, 19:5-67. By the late
`
`
`
`17
`
`
`
`HPE, Exh. 1003, p. 17
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`1990s, so-called network-attached distributed storage systems or storage area
`
`networks were developed by IBM, Sun Microsystems, and numerous others. Ex.
`
`1008 at 32-42.
`
`29. A well-known appeal of such distributed storage systems was
`
`scalability. A distributed system could store more data, and support additional users,
`
`by increasing the number of devices on the network. Ex. 1007 § 5.7 (“add[ing] a
`
`new storage server to an existing system”); Ex. 1005 at 11:19-23. In addition, the
`
`additional devices on the network did not require costly special-purpose computers.
`
`To the contrary, many systems used inexpensive, highly available components as
`
`storage and access devices. Ex. 1007 § 10 (requiring no “high-performance
`
`machines” or special-purpose hardware”); Ex. 1012 at 8 (deploying a “distributed
`
`file system . . . on an existing desktop infrastructure”).
`
`30. A high-level schematic of a distributed storage system using general
`
`purpose computers is shown below:
`
`
`
`18
`
`
`
`HPE, Exh. 1003, p. 18
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`
`
`
`Ex. 1007 at 11. As shown, the typical storage network consisted of (1) clients
`
`seeking to access storage; (2) storage servers storing the files or data objects to be
`
`accessed; and (3) a file manager managing the clients’ access to the storage servers
`
`and “the file and directory structure of the file system.” See Ex. 1007 §§ 3, 3.3, 4-
`
`4.3; Ex. 1005 at 3:61-4:11, 9:21-67; Ex. 1004 at Abst. (disclosing a “virtual file
`
`system that manages the files” and “conduct[s] file system operations” for the
`
`client), 3:3-60, 5:39-53, Figs. 11-13B.
`
`31. Notably, the storage servers on such a distributed storage system
`
`necessarily had an external interface—e.g., an interface to receive and send inputs
`
`and outputs to other external devices on the system. For example, if a client wanted
`
`to perform basic file system operations (e.g., “open” or “delete” a file) (Ex. 1004 at
`
`14:31-33, 6:67-7:2, Ex. 1005 at 8:8-19, 9:9-12), it would send a request for the file;
`
`
`
`19
`
`
`
`HPE, Exh. 1003, p. 19
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`the file manager would determine where the file is, and the request would be sent to
`
`the external input of a storage server, which would provide the requested file on its
`
`output.
`
`32. As an example, the figure below illustrates a distributed storage system
`
`in the late 1990s comprising clients (bottom of figure), storage devices (top left
`
`devices), and a file manager (top right). Ex. 1011 at 10. The storage devices are
`
`attached through an interface directly to a local area network to receive
`
`inputs/requests and provide outputs/responses. The file manager, for example,
`
`“translates client requests into commands to disks,” and the storage servers receive
`
`the command through their network interface. Ex. 1011 at 9. Storage devices in any
`
`such distributed storage area network required an interface to receive and respond to
`
`input/output requests.
`
`
`
`
`
`20
`
`
`
`HPE, Exh. 1003, p. 20
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`33. Moreover, these known distributed storage systems also adopted
`
`naming schemes to identify files or data objects stored on storage servers—a concept
`
`as old as file storage systems themselves. As just one example, systems often used
`
`resource locators “to specify the location of resources,” including files, on a network.
`
`Ex. 1009 at 5:40-49; Ex. 1004 at 5:44-46, 7:50-8:65, 26:29-54. Hostnames and
`
`pathnames, for example, were also well-known tools for identifying files at a
`
`particular network address. Ex. 1009 at 6:16-65. Naming schemes were
`
`implemented in a manner transparent to the user. Id. at 6:6-13; Ex. 1008 at 21.
`
`B. Redundancy and Selecting Storage Devices for Client Access
`34. Data redundancy or replication has been a key feature of storage
`
`systems since their inception. Data redundancy/replication improves fault tolerance
`
`and enhances reliability and availability “by making multiple replicas of each file
`
`and distributing them.” Ex. 1012, § 1; Ex. 1004 at 19:43-45; Ex. 1005 at 10:1-25
`
`(“The only way to guarantee that data is safely stored is to replicate the data . . .”).
`
`Beginning with local storage systems, architectures such as RAID (Redundant Array
`
`of Inexpensive Disks) provided configurations to replicate data across disks and
`
`recover data after disk failure. Ex. 1007 § 3.1; see generally Ex. 1010.
`
`35. Distributed storage systems in the 1990s similarly replicated data or
`
`files across networked storage devices. Ex. 1012 § 1. In describing a distributed
`
`
`
`21
`
`
`
`HPE, Exh. 1003, p. 21
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`storage system at Microsoft Research, for example, researchers explained how
`
`replication in such systems promotes “reliability by distributing multiple encrypted
`
`replicas of each file” across devices. Ex. 1012 § 7. Replication across storage
`
`devices also promotes file availability because if one device is busy, a client can
`
`access a replica on another device. Ex. 1004 at 19:43-45 (explaining that “fully
`
`redundant” network components support a “high level of service availability”),
`
`19:58-64; Ex. 1007, § 6.
`
`36. Given the multiple copies of files or data objects across a replicated
`
`system, file managers were required to select which of many file replicas to access
`
`in response to a client request. Ex. 1007 § 2.1, p. 8 (“multiple paths must exist . . .
`
`so that different paths can be used to reach different disks.”), Ex. 1004 at 11:55-62,
`
`13:5-20. In prior art systems, the file managers on the network made this selection
`
`based on several considerations.
`
`37. For example, file managers would consider the availability of storage
`
`devices and select devices to maintain performance and avoid hot spots, through
`
`implementation of load balancing and bandwidth distribution schemes. Ex. 1004 at
`
`FIG. 6 (assessing “health” and “load” of “storage nodes”), 7:11-32, 19:20-24, 19:39-
`
`56, 21:50-57; Ex. 1005 at 3:20-22; Ex. 1012, § 2.1 (“Second, the selected machine
`
`should be lightly loaded”), Ex. 1007 § 6.
`
`
`
`22
`
`
`
`HPE, Exh. 1003, p. 22
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`38. As another example, it was known (and indeed common sense) to
`
`access data from a nearby device rather than a remote device. Ex. 1012 § 2.1 (“First,
`
`the selected machine should be topologically close to the machine that is requesting
`
`the file”). Placing networked storage at multiple locations allowed the system to
`
`achieve such locality. See e.g., Ex. 1004 at FIG. 15. In these systems, the file
`
`manager located and selected the replica nearest to the client to speed up access and
`
`reduce network traffic. Ex. 1007 § 9 (“improv[ing] performance by allowing a client
`
`to access the nearest replica”), Ex. 1005 at 3:23-40; Ex. 1004 at 13:10-17. In systems
`
`considering locality, the file manager would check the client request to determine
`
`which nearby device it is associated with, which the file manager would then assess
`
`for availability and load. For example, a system may perform “geographic load
`
`balancing” by directing requests to “to the nearest storage center,” and only selecting
`
`remote storage if local storage is unavailable. Ex. 1004 at 19:15-17.
`
`VI. OVERVIEW OF THE ’082 PATENT
`A.
`Priority Date
`39.
`I have assumed the priority date of the ’082 Patent is its filing date,
`
`February 5, 2001. I have, therefore, applied this date in considering the prior art and
`
`the viewpoint of a POSITA.
`
`
`
`23
`
`
`
`HPE, Exh. 1003, p. 23
`
`

`

`Declaration of Todd Mowry, Ph.D.
`IPR2022-00290
`U.S. Patent No. 6,779,082
`B.
`The ’082 Patent Specification
`40. The ’082 Patent