`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`NETAPP, INC.,
`Petitioner,
`
`v.
`
`INTELLECTUAL VENTURES II, LLC,
`Patent Owner.
`__________
`
`Case IPR2017-00276
`Patent 6,633,945
`__________
`
`Record of Oral Hearing
`Held: March 1, 2018
`__________
`
`
`
`
`Before JEFFREY S. SMITH, JENNIFER S. BISK, and BEVERLY M.
`BUNTING, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`BENJAMIN E. WEED, ESQ.
`K&L Gates LLP
`70 West Madison Street, Suite 3300
`Chicago, Illinois 60602-4207
`(312) 781-7166
`benjamin.weed@klgates.com
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`LORI A. GORDON, ESQ.
`Sterne, Kessler, Goldstein & Fox P.L.L.C.
`1100 New York Avenue, NW
`Washington, D.C. 20005
`(202) 772-8862
`lgordon@skgf.com
`
`
`
`
`The above-entitled matter came on for hearing on Thursday, March
`
`1, 2018, commencing at 10:00 a.m. at the U.S. Patent and Trademark Office,
`600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`2
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`
`P R O C E E D I N G S
`- - - - -
`
`
`
`9:59 a.m.
`
`CLERK: All rise.
`JUDGE SMITH: Please be seated. Good morning, welcome to the
`Patent Trial and Appeal Board. We're here this morning for a hearing in
`inter partes review, case IPR 2017-00429.
`This is a case where EMC Corporation, Lenovo, and NetApp are the
`Petitioner, and Intellectual Ventures is the Patent Owner.
`I'd like to start by getting appearance of counsel. Petitioner, please
`step up to the podium and state your appearance.
`MR. WEED: Good morning, your honor. Ben Weed from K&L
`Gates on behalf of Petitioner. And with me is my colleague Erik Halverson.
`I also just wanted to note for the record that we're here I think for IPR 2017-
`00276.
`
`JUDGE SMITH: 270 -- oh, I'm sorry. My apologies. Sorry, my
`mistake, 276. The wrong file. Okay, Patent Owner, please state your
`appearance.
`MS. GORDON: Good morning, your honor, so I'm Lori Gordon
`from the law firm of Sterne Kessler Goldstein and Fox, and I'm going to be
`arguing today on behalf of the Patent Owner, Intellectual Ventures.
`With me at counsel table is Steve Peters, also from Sterne Kessler.
`JUDGE SMITH: Thank you. Both parties have given us slides that
`they want to show us during the presentation. I just want to ask both parties
`
`3
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`to keep in mind, when you reference a slide, please identify the slide by
`number so that way we can keep track of which slide you're looking at
`during the hearing. And also when we review the transcript, it makes it
`easier to review the transcript.
`Petitioner, you'll present your case first, Patent Owner will have a
`chance to respond. And then if Petitioner reserves time for rebuttal,
`Petitioner will have a chance for rebuttal. Do you wish to reserve time,
`Petitioner?
`MR. WEED: Yes, your honor. At this point, I plan to reserve 13
`minutes, please.
`JUDGE SMITH: Okay, so that'll leave you 17 minutes for your
`initial presentation. You may proceed when you're ready.
`MR. WEED: And your honor, I don't know whether it would help
`the panel, but I do have hard copies of the slides if you'd like.
`JUDGE SMITH: I don't think we need them.
`JUDGE BISK: I don't.
`MR. WEED: Good morning, Your Honors. As I mentioned, my
`name is Ben Weed from K&L Gates. And we're here today to argue a case
`that's a little bit nonstandard for IPR matters.
`And the reason why is that the primary ground of invalidity that the
`petition focused on doesn't get a lot of attention in the papers post-
`institution, especially from the patent owner.
`And I think the reason why that's the case is that the question we're
`trying to answer here today is a pretty simple question. Namely, the
`
`4
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`question is: Would it have been obvious to modify the two primary
`references to rely on a known networking protocol or scheme to connect
`nodes within those references? And particularly with regard to the first
`ground, I think the answer to that is absolutely yes.
`Another thing that I think it's important to keep in mind in this case -
`- and this was in our petition at page seven -- is that during examination of
`the 945 patent, the applicant was faced with a 112 rejection.
`And in response to that, the applicant pointed to some material in the
`specification that we'll look at, but what the applicant said to the examiner
`was, quote, The level of skill in the art is very high, thus the corresponding
`teaching can be relatively concise.
`So again, we're looking at the question of whether it would have
`been obvious to use one of a handful of known interconnection techniques
`that had known benefits -- and admittedly known deficiencies -- in the
`context of the two primary references.
`If we could flip over to Slide 2, please. As I mentioned, the 945
`patent does contain some disclosure to support the claim limitation that will
`be the focus of much of the discussion today, and that limitation regards a
`third point-to-point connection.
`On Slide 2, we've reproduced a passage from Column 7 of the 945
`patent, and in particular, the highlighted portion says, the interconnection
`between FCUs are point-to-point and fully connected. Each FCU has
`direction connection to all other FCUs.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`5
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`And then in the figure that we've got on Slide 2, which is from
`Figure 12 of the 945 patent, we've drawn a red box around the illustration of
`this interconnection between FCUs. This is the concise disclosure the
`applicant was talking about in the prosecution history, and this is the thing
`we're looking for when we're looking at the obviousness question in this
`case.
`
`If we flip over to Slide 3, we've reproduced Claim 1, which is one of
`only two challenged claims in this case. And I believe it would be fair to say
`that the parties have treated it as representative for purposes of the
`arguments.
`Now, Claim 1 has a lot of limitations in it, there's a lot of words to it.
`But if we flip to Slide 4, we see that the panel's job is actually not that
`involved in this case. On Slide 4 we've reproduced the portion of Claim 1
`that in our view is undisputed.
`The Patent Owner has not disputed that any of these limitations are
`found in the two primary references that are used in the three instituted
`grounds.
`Now, for shorthand in the course of this proceeding, we've called
`these limitations the node limitations. We talked about that, for example, in
`the petition at page nine. But really what this language is talking about is
`having a first node with some internal structure, and a second node with very
`similar internal structure.
`And the patent owner hasn't disputed that this internal structure is
`present in both Ekanadham and in Sharma. And so that's why I said before
`
`6
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`that the real dispute here is with regard to the language on Slide 5, which
`relates to the claim's third point-to-point connection.
`And now that we've seen what we're calling nodes in the petition, the
`third point-to-point connection is there to connect the nodes to one another
`so that they can exchange data.
`JUDGE BISK: Can I ask a question about this limitation? I'm a
`little confused about whether the parties want us to construe a point-to-point
`connection. And what is your take on that? Does it need to be construed,
`and if so, what should it be?
`MR. WEED: Your Honor, it's our view that it does need to be
`construed, although I will note that the construction issue doesn't play into
`the first ground in this proceeding.
`JUDGE BISK: Okay.
`MR. WEED: So if the first ground is the one that is the deciding
`ground, the claim construction disputes don't matter. The claim construction
`disputes really only matter if you're looking at a Sharma-based ground.
`So if we flip over to Slide 6, we can talk briefly about claim
`construction, again with the understanding that this is about the second and
`third instituted grounds.
`The Petitioner's proposed construction on Slide 6 is basically what
`we've highlighted in green. And it's our view that this is definitional
`language from a parent of the 945 patent.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`7
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`I think it's important to note that this doesn't talk about a third point-
`to-point connection, it's simply defining the claimed point-to-point
`connection. So to us, this is a definition of this term.
`On Slide 7, we've tried to summarize what the Patent Owner's
`position is on this issue. And what it really seems to us is that the Patent
`Owner hasn't proposed an affirmative construction. In the Patent Owner
`response at page 15, the Patent Owner basically says we don't need to
`construe this term, because under any construction, we, the Patent Owner,
`wins.
`
`But then later on in the papers, they go on to say that, well, at the
`very least, the point-to-point connection can't include a switch. So in our
`view, what the Patent Owner is asking for is a negative limitation that says
`whatever a point-to-point connection is, we don't need to concern ourselves
`with, because VivaTech tells us we don't. But it can't be a switch.
`Now, that matters because in Sharma, the connection mechanism in
`the Sharma reference on its own is called the hierarchical switch. And so I
`think that's why they want that construction. But part of the issue here is
`that the Patterson reference -- which we've also excerpted on Slide 7, Exhibit
`2002 -- belies their position.
`Patterson is a reference the Patent Owner itself put into the record,
`and while their expert testified he was familiar with the book, he didn't look
`at it for this proceeding. But what Patterson tells us is at least the Patterson's
`author's switches could be examples of point-to-point connections.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`8
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`So in our view, the construction they offer -- which is a negative
`limitation only -- isn't even supported by their own references.
`JUDGE SMITH: But it's -- I mean, the construction that we're using
`here is not broadest reasonable construction, is that right?
`MR. WEED: Correct, correct. It's a Phillips standard.
`JUDGE SMITH: It's a Phillips standard. So you know, under the
`Phillips standard, we're going to look first at the intrinsic evidence. And
`isn't their construction consistent with the intrinsic evidence which draws the
`distinction?
`MR. WEED: Well, respectfully, Your Honor, if we go back to Slide
`6, I believe that our construction is completely consistent with the intrinsic
`record, because this is definitional language. The green highlighting is
`definitional language.
`JUDGE SMITH: Right, but then what you have quoted there, that
`block quote, you've highlighted one sentence, but then the next two
`sentences go on to distinguish statically configured communication link
`from switches. So the last sentence of that quote that you have up there
`distinguishes the point-to-point connection from a switch.
`MR. WEED: Right, so the last sentence there, the subject of that
`sentence is the channel 210. And in the definitional language we've
`highlighted, what it's talking about is a point-to-point connection, more
`generically, a point-to-point connection is this.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`9
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`JUDGE SMITH: Right, and it's saying it is a statically configured
`communication link, and it differs from a switch, from a crossbar -- I'm
`reading what you have there.
`MR. WEED: Right, right, it says a channel 210 differs from most
`common types of shared or dynamically configured interconnect structures,
`such as a bus network, ring, crossbar switch network, and the like.
`JUDGE SMITH: Right.
`MR. WEED: So I still don't think, though, that the importation of a
`single one of those limitations as a negative limitation and as being the sole
`construction that's applied under a Phillips standard is appropriate in view of
`the intrinsic record.
`JUDGE BISK: What about the -- so I'm not quite as sure as you are
`that this is -- this one sentence is a definitional. What is the plain and
`ordinary meaning of point-to-point connection? Does it include a switch or
`not?
`
`MR. WEED: Well again, if we look back at Slide 7, the Patterson
`reference tells us that at least to Patterson's authors, it could include a
`switch.
`JUDGE BISK: Right, but then the Patent Owner pointed to a bunch
`of other references, including a dictionary, that seemed to imply the
`opposite.
`MR. WEED: Right. So I think the point is is that the briefing hasn't
`really developed on this point, and that's part of the problem with their
`position in this case, is they've said, we don't need to construe this term.
`
`10
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`Which, as the questions are here sort of evidencing, I think there is a need to
`construe the term.
`The only thing they've offered is that one of the things from the
`intrinsic record which is a negative limitation, applied to the channel 210
`example, is the appropriate construction.
`JUDGE SMITH: I mean you're saying negative limitation like
`there's something wrong with that. What's wrong with that? I mean, so
`what if -- I mean, the text that they're citing -- that you're citing, says it's
`distinguished from these things. What's wrong with saying it's distinguished
`from these things?
`MR. WEED: I just don't think that it's appropriate if that's the only
`construction, because we don't know what it could be. We don't know what
`it is. And especially under a Phillips world, where the construction here is
`not so much the broadest reasonable interpretation --
`JUDGE SMITH: But you're saying it is what the intrinsic evidence
`says it's distinguished from. The construction that you're offering is what
`the record says it's distinguishing the point-to-point connection from. Is that
`right? You're offering us a construction that encompasses switch, even
`though the intrinsic record distinguishes point-to-point from switch.
`MR. WEED: Well I'm offering -- we're offering a construction
`where the nature of something as a point-to-point connection is defined by
`whether or not it's a statically configured communications link. So it's not so
`much that it's not a switch, but it's rather --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`11
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`JUDGE SMITH: Is a switch a statically configured communications
`
`link?
`
`MR. WEED: I think the experts agree that it could be.
`JUDGE BISK: And I think, though, the Patent Owner also disputes
`that in Sharma that it is a statically configured.
`MR. WEED: Correct, that's right, that's right. And that's, I think, a
`different issue. That's the application of the construction issue. And so I
`think the Patent Owner, with regard to Sharma, makes two arguments.
`One is, under our proposed construction -- which is a statically
`configured communications link like on Slide 6 -- Sharma doesn't need that
`for some reasons that they argue.
`But then they also say -- more generally and more simply to them --
`because Sharma relies on a switch, if the negative limitation is the one that
`attaches, then Sharma's just -- we don't have to go any further because
`Sharma has a switch.
`JUDGE BISK: So I have one more question. You said that the
`claim construction matters for Grounds 2 and 3, Sharma by itself and
`Sharma and Hagersten.
`MR. WEED: Correct, correct.
`JUDGE BISK: Why does it matter for the Sharma and Hagersten?
`Because I thought you were looking to Hagersten for the point-to-point
`connection.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`12
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`MR. WEED: I think it matters for that ground, because if a switch
`can't be a point-to-point connection, the motivation to combine analysis
`changes a bit.
`JUDGE BISK: I see.
`MR. WEED: You're not looking at things where you've got kind of
`a genus-species arrangement, but rather set one of items and set two of
`items.
`
`JUDGE BISK: Okay.
`JUDGE SMITH: Let's say we don't agree that that point-to-point
`encompasses switch. Let's say we decline to say that.
`MR. WEED: Right.
`JUDGE SMITH: What happens then to the arguments that you're
`presenting before us? Are the claims still patentable? If we say that that
`point-to-point excludes switch, are the claims still -- do you think the claims
`are still unpatentable in that case?
`MR. WEED: Yeah, absolutely. So like I mentioned earlier, our first
`ground -- which is Ekanadham in view of Hagersten -- doesn't implicate this
`concern at all. So that ground is not affected by claim construction in our
`view.
`
`And even under the Sharma and Sharma plus Hagersten grounds, if
`the connection between the nodes can't be a switch, if that negative
`limitation is the appropriate way to construe the term, we've presented
`arguments in the reply about why the references still render those claims
`obvious. And we can look to, for example, Slide 16 on this point.
`
`13
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`As I mentioned in the beginning, there is no dispute that Sharma
`meets the individual node-specific limitations. So again, the dispute here is
`whether or not the nodes of Sharma, it would be obvious to connect those
`nodes using a point-to-point connection, whatever that means.
`In Slide 16, we've got a picture from Figure 7 of Sharma, which is
`the architecture of Sharma. And so we can see that Sharma calls the nodes
`SNP nodes and illustrates a hierarchical switch as being in between those
`SNP nodes.
`Now, the reply presented some arguments about why even if the
`construction doesn't per se exclude switches, why Sharma is still a disclosure
`of a statically configured communications link. But I think the Panel's
`question was what happens if switch is excluded.
`And so we can see that on Slide 19, for example, which again -- this
`is argument from the reply based on this claim construction position the
`Patent Owner took in the Patent Owner response.
`JUDGE BISK: So this, this wasn't in the petition anywhere?
`MR. WEED: Correct, correct.
`JUDGE BISK: Okay.
`MR. WEED: Again, this is in response to their argument that the
`construction is appropriately just, you know, a switch can't be covered.
`We're not going to say what a point-to-point connection is, but a switch can't
`be covered.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`14
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`JUDGE BISK: But I thought in the petition you had already -- you
`had kind of figured out that that was going to be their argument, because
`that's the whole point for having Hagersten, right?
`MR. WEED: Right, so the petition contemplated a scenario where
`the Patent Owner presented a construction where there might be something
`permitted to be in the middle. So it wasn't a -- what we contemplated in the
`petition was not a per se exclusion of switches, but rather the ability to have
`something more than just fiber or wire between two nodes.
`And so even with that sort of assumption --
`JUDGE BISK: Wait, I'm sorry, I'm not understanding that. So your
`position is that you thought they were going to say there could be something
`--
`
`MR. WEED: Right.
`JUDGE BISK: But not a switch, or --
`MR. WEED: No, we didn't anticipate that they would be coming
`with a position that a switch would be excluded per se.
`JUDGE BISK: Okay.
`MR. WEED: So our view was point-to-point connection, in our
`view, didn't have anything in the middle. There was just wires or fiber
`between Node 1 and Node 2.
`JUDGE BISK: Right.
`MR. WEED: But we thought, based on what they had done in the
`District Court case that perhaps they would argue something a bit broader,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`15
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`where there could be intervening logic or other structures between Node 1
`and Node 2.
`JUDGE BISK: Oh, I see.
`MR. WEED: And still satisfy the point-to-point connection. What
`we didn't think they would do or even think, you know, was a possibility
`was to say, the only construction here is that a switch can't be part of a point-
`to-point connection.
`So these arguments that we're talking about on Slide 19 are in
`response to that position from the Patent Owner. We tried to anticipate and
`weren't able to anticipate what they were going to do on claim construction.
`But what happened in the Patent Owner response is the expert for the
`Patent Owner, Dr. Smith, testified about why a hierarchical switch -- which
`is a switch -- doesn't meet the Patent Owner's construction. And so in his
`deposition we asked him, you know, tell us about a hierarchical switch.
`What is it that a hierarchical switch does.
`And he didn't have a lot to add beyond what Sharma itself says. So
`in the reply, I went back and looked at what Sharma says about a
`hierarchical switch.
`And what Sharma actually does at Column 13, lines 10 to 17, is it
`points us to another reference incorporated by reference in Sharma and says,
`this other reference -- which by the way was filed the same day as Sharma --
`is a disclosure of a hierarchical switch that is suitable for use in the
`embodiments of Sharma.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`16
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`So we went and took a look at that reference, just to see if we could
`get some more detail around this hierarchical switch, which had become a
`much more disputed issue after the Patent Owner response.
`JUDGE SMITH: You have about a minute left.
`MR. WEED: About a minute?
`JUDGE SMITH: Yeah.
`MR. WEED: What we saw in there was that there was a -- the
`incorporated by reference document VanDoren discloses what we see on
`Figure 7b on Slide 19, which is a point-to-point connection.
`VanDoren tells us that even in the context of Sharma, those of skill
`in the art understand that if we're talking about a two-node embodiment --
`which is what's claimed -- a point-to-point connection like what we see in
`Figure 7b is perfectly appropriate to still provide the functionality that
`Sharma's HS -- or hierarchical switch -- needed.
`And with that, I will reserve the rest of my time.
`JUDGE SMITH: Thank you.
`MS. GORDON: Okay. Thank you, Your Honor, and good morning.
`Again, I'm Lori Gordon from the law firm of Sterne, Kessler, and I'm going
`to be arguing on behalf of the Patent Owner, Intellectual Ventures.
`And Petitioner's counsel started their argument with a statement that
`this is a nonstandard IPR proceeding. And Patent Owner agrees, but not for
`the reasons that Petitioner's counsel stated. This is nonstandard because
`Petitioner presented two combination arguments and nowhere clearly
`articulated what combination they were proposing.
`
`17
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`They provided no notice for the Patent Owner so that Patent Owner
`could assess the viability of their combination. And as we know, one of the
`key tenets of obviousness doesn't stop with whether two references can be
`combined. The questions then becomes whether a person of ordinary skill in
`the art would have combined those references.
`And a key component of that analysis is whether the proposed
`combination would have destroyed the principle of operation or rendered the
`base reference unsuitable for its intended purpose.
`And by not giving Patent Owner any indication of the precise
`combination they were proposing, other than you can take pieces, generic
`pieces, from this reference and generic pieces from that reference and that's
`all we have to do, Patent Owner had no notice and no meaningful
`opportunity to respond to their arguments.
`And so that's why this is a nonstandard case, because Petitioner
`shifted the burden to the Patent Owner to figure out their arguments, and
`that's simply improper.
`It's also nonstandard, because what we saw and we heard here today
`is that Petitioners waited until their reply and their arguments here today to
`start to articulate what their positions are on various points. And that's
`improper, and the Board should disregard these new arguments.
`So in an IPR proceeding, we know that the burden is on the
`Petitioner to establish unpatentability. And that burden attaches with the
`petition, which is the only paper that the Patent Owner has a full and fair
`opportunity to respond to.
`
`18
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`So today and in our Patent Owner response, we focused on what the
`Petitioner said in the petition, because that was the only thing we had in
`front of us.
`And we'd like to focus the Board's attention on that today, and start
`our arguments with Ground 4, which is Sharma alone renders the challenge
`claims unpatentable. And I'd like to turn to our Slide 11.
`Again, starting with precisely what the Petitioner said in their
`petition. Petitioner told us that they're presenting Ground 4 to the extent the
`Patent Owner proposes a broader construction for point-to-point connection
`that would not require a statically configured communication link between
`two devices.
`In other words, the only reason they were presenting Ground 4 was
`to the extent that the Patent Owner proposed a different construction.
`Well, Patent Owner did not propose a different construction. So
`there's no alternate construction at issue in this case, and this should end the
`inquiry as to Ground 4, because Patent Owner does --
`JUDGE BISK: Can I ask you about that?
`MS. GORDON: Yes.
`JUDGE BISK: So I'm still a little confused about whether in the
`decision we need to discuss the construction of point-to-point connection.
`Because it seems that maybe both sides agree that a point-to-point
`connection is a statically configured communications link. But they don't
`agree on whether that includes a switch or not.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`19
`
`
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`Case IPR2017-00276
`Patent 6,633,945
`
`MS. GORDON: Your Honor, I think the Board -- the term is at
`dispute in this case. And I think the only dispute is the contours and
`applicability of the construction. So Patent Owner does not dispute that a
`point-to-point connection is a statically configured communication link
`between two devices.
`And in fact, at page 32 of our POR, we stressed that there seemed to
`be no dispute that a point-to-point connection had two different
`characteristics, that it was a direct communication link, and that it was
`between only two devices.
`And at deposition, their expert agreed with us that those were two
`characteristics of a point-to-point connection. So we think there's no dispute
`between the parties that that is the proper construction in this case.
`JUDGE BISK: So the dispute is whether a switch can be a statically
`configured communications link, or if it's always dynamic, is that?
`MS. GORDON: Yes. So the dispute is whether a switch falls within
`that construction.
`JUDGE BISK: Okay.
`MS. GORDON: And we presented various arguments that it doesn't,
`starting with --
`JUDGE BISK: Now does this matter for the -- I'm sorry, I can't say
`the name of that.
`MS. GORDON: The Ekanadham?
`JUDGE BISK: Yeah, that one.
`MS. GORDON: Yes.
`
`20
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`JUDGE BISK: Does it matter for that combination whether --
`because it doesn't seem to be a switch that's the issue in that one.
`MS. GORDON: Yeah, so this only matters for the single reference
`Sharma ground.
`JUDGE BISK: Okay.
`MS. GORDON: And as we can see from Petitioner's concession in
`the petition, it only matters if this isn't the construction that's being proposed.
`And in fact, I think there's both -- agreement on both sides that it is. So I
`think Petitioner has recognized that a switch doesn't meet this construction,
`and therefore they proposed this as an alternative.
`So for the Ekanadham and Hagersten combination, in fact for the
`Sharma and Hagersten combination, it also does not matter.
`JUDGE BISK: Okay, even for the rationale to combine?
`MS. GORDON: It doesn't, because based on the testimony of their
`expert at deposition, their combination removes the hierarchical switch. So
`it's no longer a component, to the best that we can understand their
`combination.
`JUDGE BISK: I see.
`MS. GORDON: The argument related to that combination is that by
`replacing that hierarchical switch with point-to-point connections, you have
`destroyed the principle of operation of that reference, and in fact rendered it
`inoperable for its -- not only its intended purpose, but it would no longer be
`a viable symmetric multiprocessor system.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`
`21
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`JUDGE SMITH: Before we get to that, let me just -- I just want to
`finish up on what you think the construction of point-to-point is, whether it
`excludes switches. I think -- we've read your arguments in the briefs, but I
`think the Petitioner's point is that a document that you submitted says that
`switches are point-to-point connections.
`So they're relying on evidence submitted by Patent Owner to support
`the construction of point-to-point as encompassing switches.
`MS. GORDON: Yes, Your Honor. And it is our position that the
`construction of point-to-point connection does not include switches. And I'd
`like to address Petitioner's statement related to Patterson, because they are
`mischaracterizing that Patterson reference. So I'm going to put up
`Petitioner's Slide 7.
`And so I've put Petitioner's Slide 7 on the ELMO. And what
`Petitioner has neglected to highlight for the Board is that Patterson is not
`talking about a point-to-point connection, right. So he's not saying a switch
`has a point-to-point link -- communication link between it. He's talking
`about point-to-point communication. And that is between Device A and
`Device B.
`So what Patterson is saying is if you have a switch, it makes
`communication between two points more efficient. And that's different if
`you know how a bus works, because buses work differently than switches.
`So it's not really a point-to-point communication.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`
`22
`
`
`
`Case IPR2017-00276
`Patent 6,633,945
`
`So what Patterson was stressing was point-to-point communication,
`which is very different than a point-to-point link, which is a direct link
`between two devices. And switches just are not designed to do that.
`In fact, the whole point of switches in the telecommunication
`industry was to get away from point-to-point links, because they provide
`scalability.
`JUDGE SMITH: Let me ask you -- because I don't want to take too
`much time on this, let me ask you also about Petitioner's point that if we say
`that -- if we do decide to construe point-to-point connection, and we do
`construe it to say point-to-point excludes switches, we would be importing a
`negative limitation into the claim.
`And you know, Petitioner is telling us there's something wrong with
`that because we should construe the scope of what the term is rather than
`what it is not. Do you agree with t