`EMC Corporation v. Intellectual Ventures II LLC
`
` Volume 2, Pages 58-134
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
` Case No. IPR2016-01106
` ---------------------------
`EMC CORPORATION,
` Petitioner,
`vs.
`INTELLECTUAL VENTURES II LLC,
` Patent Owner.
` -----------------------------
`
` CONTINUED DEPOSITION OF DOUGLAS CLARK, Ph.D.
` Monday, July 17, 2017, 9:02 a.m.
` WilmerHale
` 60 State Street
` Boston, Massachusetts
`
` ------- Reporter: Alan H. Brock, RDR, CRR -------
` Advanced Depositions
` 855.811.DEPO
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 58
`
`INTELLECTUAL VENTURES EX. 2021
`EMC v. Intellectual Ventures
`IPR2016-01106
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`APPEARANCES:
`
`Wilmer Cutler Pickering Hale and Dorr LLP
` Theodoros Konstantakopoulos, Esq., Ph.D.
` 7 World Trade Center
` 250 Greenwich Street
` New York, New York 10007
` 212.295.6367
` theodoros.konstantakopoulos@wilmerhale.com
` for Petitioner
`
`Knobbe Martens
` David Jankowski, Esq.
` 2040 Main Street, Floor 14
` Irvine, California 92614
` 949.760.0404
` david.jankowski@knobbe.com
` for Patent Owner
`
`ALSO PRESENT:
`James R. Hietala (by telephone)
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`59
`
`1
`
`2 3
`
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`Page 59
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
` I N D E X
`
` EXAMINATIONS
` DOUGLAS CLARK, PH.D.
` MR. JANKOWSKI 61
`
` EXHIBITS PREVIOUSLY MARKED
` 1001................................... 66
` 1002................................... 63
` 1003................................... 123
` 1004................................... 96
` 1028................................... 65
` 1029................................... 88
` 1030................................... 85
` 1031................................... 86
` 1032................................... 83
` 1033................................... 87
` 1034................................... 90
` 1035................................... 91
` 1037................................... 103
` 2013................................... 94
` 2020................................... 93
`
`No exhibits marked
`
`1
`
`2 3
`
`4
`5
`
`6 7
`
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`60
`
`Page 60
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` July 17, 2017 9:02 a.m.
` P R O C E E D I N G S
` DOUGLAS CLARK, PH.D.,
`being first duly sworn or affirmed to testify to the
`truth, the whole truth, and nothing but the truth,
`was examined and testified as follows:
` EXAMINATION
`BY MR. JANKOWSKI:
` Q. Good morning. Could you please state your
`name for the record.
` A. I'm Douglas Clark.
` Q. And you've been deposed once before in this
`proceeding; correct?
` A. Yes.
` Q. How many times have you been deposed in
`this or other proceedings?
` A. I don't know the exact number offhand, but
`it is more than ten and less than twenty.
` Q. And you're familiar, then, with the rules
`that are associated with depositions; correct?
` A. Yes.
` Q. And you're testifying as if you were in a
`court of law or in this context, I guess, in
`connection with a proceeding of a board of the
`Patent Office. You understand that; correct?
`
`61
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 61
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` A. Yes.
` Q. And what did you do to prepare for this
`deposition?
` A. Let's see. I began preparing at home an
`hour or two a day roughly once I learned that there
`was going to be this deposition, reviewing my two
`declarations and the other -- the patents, the
`documents -- the documents involved in the case.
` And then I came to Boston on Saturday
`and spent half a day with Dr. Konstantakopoulos, on
`my right, and another attorney, Peter Dichiara. And
`I spent most of yesterday with Dr.
`Konstantakopoulos.
` Q. And is there any reason you cannot provide
`full and complete truthful testimony today?
` A. There is none.
` Q. And I'm going to apologize in advance for
`the next couple of questions, but I ask them in
`every deposition that I take. Have you ever been
`accused or convicted of a crime?
` A. No.
` Q. Have you ever been disciplined in any way
`by an employer?
` A. No.
` Q. Now, your first declaration in this case,
`
`62
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 62
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`in this proceeding, relied on an interpretation for
`the term "switch fabric." Is that correct?
` A. I do not quite understand the question. An
`interpretation?
` Q. A claim construction.
` A. Yes.
` Q. And was that your own claim construction,
`or was it provided to you?
` MR. KONSTANTAKOPOULOS: Objection.
` A. In the first declaration, I cannot recall
`the details of that. I believe there were two. I
`would like to see the report, just to refresh my
`memory.
` Q. Your first declaration?
` A. Yeah.
` Q. Sure. Let me hand you your first
`declaration. It was previously marked Exhibit 1002.
` (Exhibit 1002 previously marked for
`identification.)
` A. That was provided to me.
` Q. And as you sit here today, do you have an
`interpretation for the term "switch fabric" that you
`come up with on your own?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I did not come up with any interpretation
`
`63
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 63
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`on my own.
` Q. And in your latest declaration, your reply
`declaration, did you come -- or you're also relying
`on a construction for "switch fabric"; correct?
` A. Yes.
` Q. And again, is that a claim construction
`that you come up with yourself, or is that provided
`to you?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That one was provided by the Board.
` Q. So in your second declaration, you're
`relying on the construction provided by the Board;
`correct?
` A. Yes.
` Q. And by "the Board," we're talking about the
`Patent Trial and Appeal Board of the U.S. Patent and
`Trademark Office; correct?
` A. It was those people.
` Q. For completeness, let me give you your
`second declaration. I'm sure we're going to be
`looking at it today. So let me hand that to you so
`that you have it accessible. This is previously
`marked Exhibit 1028, entitled Reply Declaration of
`Douglas W. Clark regarding U.S. Patent No.
`6,516,442?
`
`64
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 64
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` (Exhibit 1028 previously marked for
`identification.)
` Q. And let me ask you the same kind of series
`of questions for the term "channel." Is it accurate
`that you were provided a claim construction for the
`term "channel" in connection with your first
`declaration in this proceeding?
` A. Yes, it's correct; that is correct.
` Q. And that isn't the definition of "channel"
`that you came up with yourself; correct?
` A. That is correct.
` Q. And how about the term "channel" in your
`second declaration, your reply declaration, Exhibit
`1028? Is that your own construction that you apply
`for that declaration?
` A. Again, I believe it is the Board's
`construction, but let me just check.
` Q. Well, Dr. Clark, in the interest of time:
`As you sit here, do you know whether you did? I
`have a lot of questions I need to get to.
` A. I believe I -- I cannot find a crisp
`call-out of the Board's construction, but that's the
`one I used in my reply declaration.
` Q. Have you ever formed your own independent
`view of what the claim term "switch fabric" from the
`65
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 65
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`'442 patent means, independently of what's been
`provided to you?
` A. I have not.
` Q. And have you ever formed your own
`independent view of what the claim term "channel"
`means in the '442 patent claims, independently of
`what's been provided to you?
` A. I have not.
` MR. KONSTANTAKOPOULOS: Objection.
` Q. Let me provide you with the patent, the
`'442 patent, previously marked Exhibit 1001.
` (Exhibit 1001 previously marked for
`identification.)
` Q. And you're aware that one benefit that the
`'442 patent is trying to achieve is that of
`improving scalability; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I have not considered that question, and I
`don't have a ready answer.
` Q. Okay. If you look at the first page of
`Exhibit 1001, the title of the patent reads Channel
`Interface and Protocols for Cache Coherency in a
`Scalable Symmetric Multiprocessor System. Did I
`read that right?
` A. Yes.
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`66
`
`Page 66
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` Q. And so right in the title you can see
`there's a reference to "scalable" as being part of
`what the patent is directed at. Would you agree?
` A. I do not quarrel with that as the title.
` Q. And let's turn to column 1 of the patent.
`And that presents as part of it a background. I'm
`going to start reading at line 36 of column 1. It
`starts off with the words, "A first problem." Do
`you see that?
` A. Yes.
` Q. It says, "A first problem with the above-
`described traditional SMP system is that the serial
`availability of the bus limits the scalability of
`the SMP system." Did I read that correctly?
` A. Yes.
` Q. And now I'm going to skip down a couple of
`lines, about line 41, where it starts with "What is
`needed." Do you see where I am?
` A. Yes.
` Q. I'm going to read from there from the
`patent. It says, "What is needed is an SMP system
`architecture that provides greater scalability by
`permitting concurrent use of multiple buses, while
`still providing a system serialization point to
`maintain strong transaction ordering and cache
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`67
`
`Page 67
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`coherency." Did I read that correctly?
` A. Yes.
` Q. So again, you would agree this is showing
`that a goal of the patent is achieving improved
`scalability from prior art systems?
` A. I don't quarrel that that's what it says.
` Q. And as you sit here, do you have a sense
`whether the claim constructions that you're relying
`on in your reply declaration are consistent with the
`stated goal of the patent of improving scalability?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I do not have an opinion about that.
` Q. And wouldn't you agree that one way of
`promoting greater scalability in a system such as
`taught by the '442 patent is to use a switch
`architecture between simultaneous pull processors
`and shared memory that is nonblocking?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Could you ask that again?
` Q. Sure. Would you agree that one way of
`promoting greater scalability in a system such as
`taught by the '442 patent is to use a switch
`architecture between multiple processors and shared
`memory that is nonblocking?
` MR. KONSTANTAKOPOULOS: Objection.
`
`68
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 68
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` A. I have not considered that question, and I
`don't have a view.
` Q. Now let's turn to column 3 of the patent.
`If you look down at line 23 of column 3, I'm going
`to read the statement which is there. Starting --
`this is partway through the line. It starts with
`the words "The FCU contains." Do you see where I
`am?
` A. Yes.
` Q. I'm going to read that. It says, "The FCU
`contains the simultaneous switch matrix (SSM) data
`flow switch which supports nonblocking data
`transfers." Did I read that correctly?
` A. Yes.
` Q. And you're familiar with the patent, so
`you're familiar with what is being referred to here
`as the FCU; correct?
` A. Yes.
` Q. And this is also shown in the figures of
`the patent as well; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I'm sure it is, but let me just look. FCU,
`figure 2 --
` Q. Actually, that's perfect -- figure 2 shows
`the FCU as element 220; correct?
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`69
`
`Page 69
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` A. Yes.
` Q. And that's described as nonblocking;
`correct?
` MR. KONSTANTAKOPOULOS: Objection.
` Q. It's characterized as nonblocking right at
`the excerpt that I read earlier, at column 3
`starting at line 23?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I hesitate only because that passage seems
`to be crediting the SSM data for the switch and not
`the FCU.
` Q. But the SSM is within the FCU; correct?
` A. In this paragraph, yes.
` Q. Right. The SSM is shown in figure 3 of the
`patent; correct?
` If it helps you, why don't you turn to
`Page 9 of your first declaration, Exhibit 1002. And
`you provide a nice colorized picture of figure 3.
` A. What page?
` Q. Page 9.
` A. I call that the switch fabric. I was just
`looking to see whether there was a reference in the
`specification to -- for the name of the thing inside
`the FCU.
` Q. Okay, well, let's --
`
`70
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 70
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` A. Which is -- there's not a box for that,
`actually.
` Q. I'm fine using "switch fabric." The switch
`fabric that's shown and colored in red on Page 9 of
`your first declaration, that's taught by the patent
`as being nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. And figure 4 of the patent teaches a
`variation on the embodiment of figure 2; correct?
` A. Yes.
` Q. And figure 4 also says in it expressly that
`the switch matrix is nonblocking; correct?
` A. Yes.
` Q. And figure 6 of the patent shows another
`view -- another embodiment -- I'm sorry. Strike
`that.
` Figure 6 shows another view of the
`embodiment of figure 4; correct?
` A. Yes.
` Q. And so the FCU disclosed in figure 6 has a
`switch fabric that is also nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That is a reasonable conclusion.
` Q. So so far we've looked at two embodiments,
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`71
`
`Page 71
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`a figure 2 embodiment and a figure 4 embodiment,
`both of which are disclosed as having a switch
`fabric that is nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. Now, figure 7 of the patent is a drawing
`that includes a number of different application-
`specific variations on the embodiment of figure 4;
`correct?
` A. Yes.
` Q. And each of the configurations shown in
`those four figures of figure 7 include a switch
`fabric that is nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. There is -- so all show an FCU. B somehow
`shows the FCU in a box, which I don't understand
`without further study. A shows a simple system with
`just one memory control unit, MCU. And so it's not
`clear that blocking is even a consideration for it.
` Q. But the patent states that each of these
`embodiments is an application-specific variation on
`the embodiment of figure 4, which is nonblocking;
`correct?
` A. Yes.
` Q. So so far we've looked at six embodiments,
`
`72
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 72
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`all of which are disclosed as having a switch fabric
`that is nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I do quarrel with 7a. But with the others
`I do not.
` Q. And now figure 9 shows the same FCU as
`figure 6. Would you agree?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I would -- that seems reasonable, but I
`want to read the specification more closely to see
`whether this FCU has special qualities. I note that
`figure 9 shows -- it shows explicitly the Level 2
`caches for the CPUs, and I don't know whether that
`would make a difference in the nonswitch part of the
`FCU.
` Q. Let me direct your attention to column 5,
`starting at line 59. Why don't I just read -- I'm
`sorry, I think it starts at line 58. I'll start
`reading the very end of line 58 of Column 5. It has
`the word "the." Do you see where I am?
` A. Yes.
` Q. It says, "The embodiments of figures 9 and
`10 are respective variations of the embodiments of
`figures 6 and 7, with adaptation for the use of the
`integrated CPU/CCU of figure 8." Do you see that?
`
`73
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 73
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` A. Yes.
` Q. And so does that answer your question as to
`whether we're dealing with the same FCU that we're
`dealing with in figures 6 and 7?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I can't state with confidence that there is
`not something else that says, "Oh, by the way, this
`FCU has an additional feature to deal with these
`Level 2 caches" or something like that.
` Q. But as you sit here, you're not aware of
`the patent referring to any changes in the FCU
`between figures 6 and 7 and figures 9 and 10?
` A. I am not.
` MR. KONSTANTAKOPOULOS: Objection.
` Q. And figure 10, by the way, has four more
`embodiments in it; correct?
` A. Right. These are -- it's sort of like
`figure 7, with the caches, Level 2 caches, called
`out.
` Q. Right. And so so far we've looked at 11
`embodiments from the patent, all 11 of which are
`disclosed as having a switch fabric that is
`nonblocking; correct?
` CROSS-EXAMINATION
` A. That is reasonable except for my earlier
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`74
`
`Page 74
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`proviso about the one with only one memory
`interface.
` Q. Okay. Now, figures 11 and 12 show the use
`of an FCU that's described as using an industry-
`standard switching fabric; correct?
` A. Yes.
` Q. And in figures 11 and 12, those embodiments
`also have a switch fabric which is nonblocking;
`correct?
` A. That also seems reasonable.
` Q. And subject to the proviso that you
`mentioned a moment ago, we've looked at 13
`embodiments in the patent, all of which have a
`switch fabric that is nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I accept your count, yes.
` Let me back up a tiny bit. The figure
`11 and figure 12 show this industry-standard I/O
`fabric, the NG I/O fabric. And we learn nothing
`about whether it is itself a nonblocking fabric from
`the patent.
` Q. But the FCU shown in figures 11 and 12
`would be a nonblocking switch fabric; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That seems reasonable, assuming that the
`
`75
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 75
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`FCU is the same FCU that it's always been.
` Q. And as you go from the embodiment of figure
`11 to the embodiment of figure 12, the patent
`describes going from a four-way embodiment to a
`16-way embodiment; correct?
` A. Okay, so I have to go back to the previous
`answer about that. I'm reading what this is. It's
`a 16-way, because while there are four ports, four
`processor ports to the switch, CCU 0 through CCU 3,
`on the other side of each of those is a bus with
`four processors.
` And so this entire -- the switch -- the
`switching fabric of this entire system includes that
`bus. And so that's not nonblocking.
` Q. I'm referring to the FCU portion of figure
`11 and figure 12. That's nonblocking; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I actually think that doesn't make sense.
`I mean, if you consider the thing in isolation, it
`is fair to say that that's nonblocking. But in use
`it would not be nonblocking.
` Q. But each of the processors connected to the
`FCU shown in figures 11 and 12 could communicate
`simultaneously with each of the memory interfaces
`connected to the FCU.
`
`76
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 76
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` MR. KONSTANTAKOPOULOS: Objection.
` Q. Correct?
` A. No, not correct.
` Q. Let me put it another way. Doesn't the FCU
`have the same architecture shown in figure 3, in
`what you have colored as red on Page 9 of your first
`declaration?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That is a 4-by-4 crossbar, and here we have
`four CCUs and four MCUs. So it's reasonable to say
`that inside there is the same switch.
` Q. Now let's shift over and talk about the
`memory device that's recited in claim 1 of the '442
`patent. Okay?
` A. All right.
` Q. Now, is it your opinion that the memory
`device recited in claim 1 is one of the elements
`that must be present to satisfy the requirements of
`claim 1?
` A. Yes.
` Q. And why is that your opinion?
` A. I'm -- I hope I'm not missing some
`subtlety. But there it is in the claim limitation.
`The memory interface is configured to exchange the
`data with a memory device. So how do you -- how can
`77
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 77
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`you satisfy claim 1 without having a memory device?
` Q. Well, because you can have a memory
`interface without having the memory device, because
`the memory interface is configured to exchange data
`with the memory device. Isn't that one way you can
`interpret claim 1?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That seems very strange to me.
` Q. Well, let's shift over to claim 11, if
`you'd look at column 37 of the patent. Let me read
`that to you. Claim 11, "The shared-memory
`multiprocessor system of claim 1 further comprising
`microprocessors and the memory device." Do you see
`that?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. Now, does that cause you to reconsider your
`opinion on whether a memory device needs to be
`present in claim 1?
` A. Not really. I don't actually understand
`what -- how that claim fits in. But if I go back to
`claim 1, it starts, "A shared-memory multiprocessor
`system comprising." Now, anyone would say that
`system has a memory device, at least. It can't have
`one of those things without the thing -- the shared
`78
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 78
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`memory's a memory. You can't do without it.
` Q. Let's shift over to your reply declaration,
`Exhibit 1028, and let's turn to Page 4. And here
`you present your opinion on some intrinsic evidence
`relating to the Board's construction of "switch
`fabric." Is that correct?
` A. Yes.
` Q. And part of the arguments that you make
`relate to the fact that the claim refers to a memory
`device. Is that correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. And you provided your opinion that the
`recitation of a memory device in claim 1 covers a
`single memory source, such as a single SDRAM. Is
`that accurate?
` A. I don't agree that a single SDRAM is
`necessarily a memory device. I would need more
`details.
` Q. But you'd agree that a single memory device
`can consist of multiple memory modules, such as two
`or four SDRAM modules; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. And as a technical matter, a single memory
`
`79
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 79
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`interface can communicate with multiple memory
`modules, such as two or four SDRAM modules; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. That is possible, yes.
` Q. Would you agree that claim 1 of the patent
`reads on the preferred embodiments of the patent?
` MR. KONSTANTAKOPOULOS: Objection.
` A. Yes.
` Q. And if we refer back to Page 9 of your
`first declaration, the one with your colorized
`version of figure 3, this is an example that you
`would agree reads on figure 1 of the patent;
`correct? This embodiment of figure 3?
` A. I'm sure you meant claim 1?
` Q. I'm sorry. Correct -- that it reads on --
`that claim 1 reads on the embodiment shown in figure
`3.
` A. My answer is approximately yes. But, of
`course, if you did have but one memory device, your
`data switch would be remarkably simpler and would
`only have in effect one row that connected to a --
`through the various interface devices to the memory.
` Q. Why can't the data device simply be the
`four SDRAM modules shown in figure 3 on Page 9 of
`your first declaration?
`
`80
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 80
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` MR. KONSTANTAKOPOULOS: Objection.
` A. I don't see it that way. Those are
`components of the memory system which is reached in
`four independent ways in this figure, from
`connections to the 4-by-4 switch. That suggests to
`me that there are four things that may be
`independently accessed, and I would call that four
`devices. I wouldn't count the chips; I would call
`it the functional device that the processor is
`speaking to.
` Q. But as a technical matter, a memory device
`can have within it four independent SDRAMs such as
`elements 1300 through 1303 of figure 3 of the '442
`patent; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. So not looking at the '442 patent, if your
`question is can a device have multiple SDRAM chips,
`certainly.
` Q. And so the four modules shown there, the
`four SDRAM modules shown in figure 3, can constitute
`a memory device; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. I do not see it that way. I think they are
`four memory devices. And I'm not sure that we're
`right to say that because they're called SDRAM that
`81
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 81
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`they are one SDRAM. I think it's a module of SDRAM.
`We could check the specification.
` Q. But as a technical matter, a memory device
`by itself cannot be -- is not precluded from having
`multiple separate SDRAM modules in it. You'd agree
`with that?
` MR. KONSTANTAKOPOULOS: Objection.
` A. So in -- again, outside the context of the
`patent, as I said before, a memory device can have
`multiple physical devices -- chips, modules,
`whatever -- and still be one memory device. This
`picture shows four memory devices.
` Sorry, I was pointing to figure 3.
` Q. Right, figure 3. I mean, you yourself in
`your first declaration referred to these four SDRAM
`modules collectively as, quote, "the memory." Do
`you recall that?
` A. Would you draw my attention to a paragraph?
` Q. Sure. In Paragraph 30, at the very bottom
`of Page 9, it's the very last sentence, so it
`carries over to Page 10. Why don't I start reading.
`See the word "the"?
` A. I see, yes.
` Q. It says, "The memory 1300 to 4,
`communicates the memory interfaces MCUs 230, yellow,
`82
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 82
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`which exchange data packets over another set of
`channels 114, orange, with switch interfaces 3108,
`green." Did I read that correctly?
` A. Yes.
` Q. So you yourself are referring to the memory
`collectively here; correct?
` A. I am -- there is in black and white, I say
`"the memory," and nominate the four SDRAM modules.
`But I'm not talking about memory devices. That's
`the memory of the system.
` Q. And the memory of the system can be
`provided by a single memory device even when it has
`four SDRAM modules; correct?
` MR. KONSTANTAKOPOULOS: Objection.
` A. So I think we've done that already. As a
`technical matter, a memory device can have multiple
`SDRAM gizmos, chips, modules, whatever, and be a
`single device. Figure 3 shows four such devices.
` Q. Let's start looking at some of the
`extrinsic evidence associated with your reply
`declaration and the construction of "switch fabric."
`I'm going to hand you previously marked Exhibit
`1032.
` (Exhibit 1032 previously marked for
`identification.)
`
`83
`
`ADVANCED DEPOSITIONS
`www.advanceddepositions.com | 855.811.3376
`
`Page 83
`
`
`
`Douglas Clark, Ph.D., Vol. II, 7/17/2017
`EMC Corporation v. Intellectual Ventures II LLC
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` Q. I'm going to ask you: Do you recognize
`this exhibit?
` A. Yes.
` Q. When did you first see this document,
`Exhibit 1032?
` A. I cannot recall whether I found this or the
`lawyers found this. It would have been -- it would
`have been -- the declaration was in June, was June
`1, so probably in May, possibly April.
` Wait a second. I'm sorry. You said
`when did I first see this document?
` Q. Correct.
` A. So it might be that I own this document. I
`can't remember. I have a few of these on my
`bookshelf. So I would have seen it, you know, years
`ago.
` Q. And that was one of my -- you're
`anticipating one of my questions, was whether you
`have the actual book, the Computer Desk
`Encyclopedia. With this particular example, Exhibit
`1032,