`UNITED STATES PATENT AND TRADEMARK OFFICE
`______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________
`
`TALARI NETWORKS, INC.,
`Petitioner,
`
`V.
`
`FATPIPE NETWORKS INDIA LIMITED,
`Patent Owner.
`______________
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`______________
`
`Record of Oral Hearing
`Held: August 14, 2017
`______________
`
`
`
`
`Before STACEY G. WHITE, MICHELLE N. WORMMEESTER, and
`CHRISTA P. ZADO, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`
`APPEARANCES:
`
`FOR THE PETITIONER:
`
` THOMAS F. FITZPATRICK
` ANDY H. CHAN
` CHARLES F. KOCH
` Pepper Hamilton LLP
` 333 Twin Dolphin Drive, Suite 400
` Redwood City, California 94065
` 650-802-3600
` fitzpatrickt@pepperlaw.com
` chana@pepperlaw.com
` kochc@pepperlaw.com
`
` FOR THE PATENT OWNER:
`
`SAMEER GOKHALE
` ROBERT MATTSON
` Oblon, McClelland, Maier & Neustadt, LLP
` 1940 Duke Street
` Alexandria, Virginia 22314
` 703-413-3000
` sgokhale@oblon.com
` rmattson@oblon.com
`
`
`
`
`The above-entitled matter came on for hearing on August 14, 2017,
`commencing at 10:00 a.m. at the U.S. Patent and Trademark Office, 26 S.
`4th Street, San Jose, California.
`
`
`
`
`
`
`
`
`2
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`
`
`P R O C E E D I N G S
` Monday, August 14, 2017 - 10:00 a.m.
` ---oOo---
` JUDGE ZADO: Please be seated. Good morning,
`everyone.
` ATTORNEYS: Good morning.
` JUDGE ZADO: This is the consolidated oral hearing
`in IPR 2016-00976 and IPR 2016-00977 between petitioner
`Talari Networks, Incorporated, and patent owner FatPipe
`Networks India Limited involving challenged claims of U.S.
`Patents 6,775,235 and 7,406,048.
` Judges White and Wormmeester are joining us
`remotely from our Dallas and Alexandria hearing rooms
`respectively, so please speak into the microphone at the
`podium and identify any exhibits or demonstratives that you
`are using so that they can be sure to pull them up.
` Each party will have 90 minutes of total argument
`time. Petitioner will go first and present its case with
`regards to the instituted claims. Petitioner may reserve
`time for rebuttal.
` Thereafter, patent owner will argue its opposition
`to petitioner's case, and if there is any rebuttal from
`petitioner, we will hear it after patent owner's opposition.
` Will petitioner's counsel please introduce
`yourself.
` MR. FITZPATRICK: Good morning, your Honor. Tom
`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
`25
`26
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`Fitzpatrick from Pepper Hamilton on behalf of Petitioner
`Talari.
` With me, by the way, is Andy Chan, my colleague
`from Pepper Hamilton, and Chuck Koch.
` JUDGE ZADO: And would petitioner like to reserve
`time for rebuttal today?
` MR. FITZPATRICK: We would, your Honor. I would
`like to reserve a half hour.
` JUDGE ZADO: Will patent owner's counsel please introduce
`yourselves.
` MR. MATTSON: Good morning, your Honor. Robert
`Mattson with the Oblon firm for patent owner FatPipe.
`Presenting for FatPipe today will be my partner Sameer
`Gokhale.
` JUDGE ZADO: Thank you. We're ready to begin
`whenever petitioner is ready.
` MR. FITZPATRICK: Thank you, your Honors.
` Thank you, your Honors and may it please the
`Board -- can you hear me?
` JUDGE WHITE: I can hear you.
` JUDGE WORMMEESTER: I can hear you.
` MR. FITZPATRICK: Thank you.
` May it please the Board, we'll go straight to slide
`2. I don't believe that the -- thank you. We'll go straight
`to slide 2 where you will notice that we have a table of
`contents of presentation for our presentation.
`
`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
`26
`
`4
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
` Each presentation is rather extensive, and, your
`Honor, we do have a separate physical copy for your Honor.
`May I approach?
` JUDGE ZADO: No, thank you. I have it
`electronically. Thank you very much.
` MR. FITZPATRICK: Each presentation, as extensive
`as they are, the slides are intended to provide us a means to
`address any questions you may have. We plan on focusing on a
`very small subset of these slides today.
` And before we dive into those slides, I think it's
`important that we describe the context of what we're talking
`about today.
` First is the level of skill in the art. It's
`undisputed that the level of skill in the art in this matter
`is relatively high. It's a Bachelor of Science in computer
`science, computer engineering, electrical engineering or
`equivalent field as well as at least two years of academic or
`industry experience in any type of networking field.
` By the way, that's Talari's view. FatPipe's level
`of skill for a person of ordinary skill is a bit broader to
`accommodate its expert, but, again, it's similar, and, again,
`it's relatively high.
` And this is important because it highlights what
`was known and what is not in dispute today.
` Routing traffic between disparate parallel
`networks, that was known. That's not in dispute.
`
`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
`26
`
`5
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
` Frame relay on the one hand and VPN on the other
`constituting disparate parallel networks, that's not in
`dispute.
` Frame relay as an example of a CO network, a
`connection oriented network, and VPN as an example of a CL or
`connectionless network, again, not in dispute.
` All of these things are well-known, even according
`to the 235 and 048 patents, in the prior art and conceded by
`FatPipe's expert, Mr. Williams.
` It's for those reasons that FatPipe does not
`contest the prior art as to many of the claimed elements of
`the 235 and 048 patents, and that's why I would like to
`address now the ones that are disputed starting with the
`dispute over per-packet basis.
` If we could go to slide 77.
` The per-packet basis issue applies to claims 4,
`that's an apparatus claim, and claim 9 of the 235 patent, and
`claim 7 and 19 of the 048 patent.
` Now, our briefing on this issue is comprehensive
`with its evidence and argument, and that's because the
`intrinsic record is clear about what the patent means when it
`says "per-packet basis."
` Karol plainly discloses, and FatPipe previously
`told Talari, the District Court and this Board that no
`construction was necessary.
` But now FatPipe seeks not only to construe the
`
`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
`26
`
`6
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`phrase that we see here on per-packet basis, but then to add
`a series of limitations that are not supported by the
`intrinsic record but also to construe their own construction.
` If we turn to slide 78, this construction fails on
`its face because these words "for each packet makes a
`discrete choice" appear nowhere. They are not in the claims.
`They are not in the spec. They are not in the file history,
`and FatPipe introduces this term "discrete choice." As
`opposed to an indiscrete choice?
` It's for this reason that FatPipe requires a
`construction of this construction adding numerous -- at least
`three limitations that likewise have no basis in the
`intrinsic record.
` So there is these three improper limitations. The
`first -- and we'll take them in turn, but the first is
`FatPipe construes its construction to exclude the patent's
`own embodiment in which packets are routed based on a single
`selection -- that is a quote -- based on a single selection
`that applies to multiple subsequent patents, a situation
`where the decision is made to route packet A down 101 and
`packets B, C, and D also go down 101 even though 280 -- these
`are highways here locally that I'm sure some of us took to
`get here today, your Honors.
` So that's one. It would exclude it in embodiment.
`The second --
` JUDGE WHITE: Let's talk about that for a second.
`
`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
`26
`
`7
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`When I look at the specification of the 235 in column 14,
`there's this disclosure column 14 around line 44 or so.
` MR. FITZPATRICK: 44 to 46, your Honor?
` JUDGE WHITE: Yes. It says the past selecting step
`908 may be performed once per packet or given selection may
`pertain to multiple packets.
` MR. FITZPATRICK: That's exactly where we were
`going, your Honor.
` JUDGE WHITE: So I'm looking at that, and I see
`that there are it looks like two separate ways disclosed
`there. Once per packet which seems to line up roughly with
`the construction that patent owners putting forward or that
`multiple packets which is the construction which is what
`you're trying to encompass as well.
` Why is that we can't look at the claim language and
`say that they made a choice to claim the once per packet
`embodiment in claim 4?
` MR. FITZPATRICK: Because that's not what's
`described in the figure that that reference is discussing,
`your Honor.
` If we look at the totality of figure 9 -- this is
`critical. If you look at the totality of figure 9, it shows
`how -- what they're trying to do to rewrite these claims.
` So when you look at figure 9, the first thing that
`you see and -- if we can go to slide 80 for a second.
` If you go to slide 80, it shows that according to
`
`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
`26
`
`8
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`the 235 patent, figure 9 illustrates methods -- this is at
`the top -- methods of the present invention, and then again
`we cite the exact reference that your Honor pointed out, and
`it's clear from the specification and both the left and right
`sides of figure 9 that this figure 9 is an embodiment in the
`invention.
` The specification states one example is the
`inventive method in which load balancing or reliability
`criteria cause an initial path selection to be made, 908, and
`then a loop occurs in which multiple packets are received,
`904, and then sent, 918, over the selected path without
`repeating the select -- the selecting steps, 908, for each
`receive, 904, send, 918, pair.
` These descriptions amplify that when you look at
`figure 9 you are not limiting your gaze to half of figure 9.
`You're not looking at it and just saying, well, as FatPipe
`wants you to do which is to say, well, they don't mean what
`they say in claim 4. They don't mean what they say in figure
`9 or in column 14, 44 through 46 or column 16, 22 through 27,
`and instead only half of figure 9 represents the invention.
` That's contrary to the intrinsic record. It's
`contrary to common sense when you look at any figure in a
`patent, and nowhere in the specification does it say, well,
`there is a red half on the left and a blue half on the right
`as FatPipe depicts on page 14 of its patent owner's response.
` And nowhere does the specification say that
`
`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
`26
`
`9
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`per-packet basis is that red half on the left. Right?
` And that figure 9's box 904 to 918, that's never
`described as anything other than selecting between network
`interfaces on a per-packet basis.
` So, for the same reason we talk about a round
`robin, and is that on a per-packet basis, and the answer
`would be yes because in a round robin for each packet a
`network interface path is chosen.
` So I think that looking at the figure as a whole
`and the specification as a whole there's not this carving,
`this carving this crevice in claim construction that you
`would need to do to do what FatPipe is seeking from your
`Honors.
` JUDGE WHITE: But are we not reading in things from
`the specification a little too much if we say that this
`figure defines what the claim has to be limited to?
`Especially when there is that language that says, you know,
`what was it, one packet or a group of packets?
` I'm paraphrasing here.
` Should we not look at this figure as just one
`particular embodiment and that it discloses more than one
`embodiment in the specification?
` MR. FITZPATRICK: So, your Honor, I'm not trying to
`limit the claims to this embodiment. What I am trying to say
`is that what FatPipe is doing is it's excluding its own
`embodiment. It's saying that you can't have a situation
`
`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
`26
`
`10
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`where one decision is made for a packet, and that applies to
`multiple subsequent packets. That's what they are saying.
` So when you look at their constructions, you
`remember, the first time they said it doesn't need a
`construction. The second time they said, well, now we need a
`construction, and this is what it means.
` Then they say, well, it needs this discrete choice,
`and then they tell you what it means to have this discrete
`choice, and this discrete choice that they're saying is
`required would mean that you cannot have a situation that's
`described in this embodiment.
` So FatPipe is the one that's trying to read out
`this claimed embodiment, this embodiment.
` JUDGE WHITE: I guess the question coming to my
`mind is the word "per" so specific that it focuses on one of
`the embodiments? Is it -- it's not that they are trying to
`burden the language. That's what the language naturally
`means when they say per packet.
` MR. FITZPATRICK: I'm not -- I may not be following
`your Honor.
` JUDGE WHITE: Let me rephrase that. I think what
`I'm getting at is the language "per-packet basis" so specific
`as to exclude this other embodiment or is it that patent
`owner is trying to make the term "per packet" that specific?
` MR. FITZPATRICK: I --
` JUDGE WHITE: Is the term "per-packet basis," is it
`
`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
`26
`
`11
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`that narrow on its own or are they trying to make it?
` MR. FITZPATRICK: I think they're trying to make
`it, your Honor. Now I absolutely understand where the Board
`is going, your Honor is going.
` That would in a sense say the per-packet basis
`requires a network path selection be made for each packet,
`and this embodiment clearly says that's not the case. Right?
` What they're importing into that claim -- into that
`claim construction is the frequency at which the network path
`decision is being made. That whenever there is a packet
`being received by their controller, we're assuming it is by
`their controller, there is a decision tree that that network
`path selection step has to be repeated, and that's not the
`case and that would rule out this embodiment from figure 9
`and from the column 14 and column 16 that the court
`referenced -- that your Honor referenced.
` JUDGE WHITE: And I think I'm looking also at the
`language. It says language -- the language in column 14 is
`may be performed -- may be performed once per packet, and
`here again in the claim language we're getting per packet.
` That seems to line up in a way as if they are
`trying to claim what is here in the spec, just the once per
`packet.
` MR. FITZPATRICK: But it would rule out the or a
`given selection may pertain to multiple packets.
` JUDGE WHITE: But don't they have the ability to
`
`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
`26
`
`12
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`claim one or the other if that's what they're trying to
`claim?
` MR. FITZPATRICK: Not if it would rule out that
`embodiment, and that's what would happen. You would
`essentially erase figure 9 in its entirety. You couldn't
`have figure 9, and you couldn't have this embodiment because
`it would be excluded if per-packet basis means each and for
`every single packet this decision tree is being made, is
`being run through.
` Your Honor, I think it also -- our reading --
`actually, well, FatPipe's reading is consistent with the
`other things that it's trying to do. So if I can go to those
`other limitations that would go hand in glove with their
`exclusion of this embodiment.
` FatPipe construes its construction such that the
`discrete choice, this is the second improper limitation, it's
`discrete choice or distinct selecting cannot be based on a
`precomputed route.
` Now, if we turn to slide 84, now, if it was
`precomputed, why does Karol say a determination is made?
`It's simply not right. The plain disclosure in Karol says
`for each packet at the CL-CO gateway that a determination is
`made if that packet should be carried by CO network 160.
` That's not a precomputed route.
` Karol also says the determination of which network
`path to select is based on service requirements and that --
`
`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
`26
`
`13
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`quote -- the traffic situation in the CL and CO networks to
`evaluate whether a particular packet may be redirected.
` And that's -- that's directly from Karol. Because
`the determination is based on the traffic situation, the
`determination could not be precomputed.
` I thought I was going to take 280 coming down
`today. 280 looks congested. I take 101.
` Again, this is the same example that we used
`earlier.
` That's something I -- that's not precomputed.
`Traffic changes over time. So to read FatPipe's construction
`of discrete choice into these claims, they have no intrinsic
`record to support any of these claims, you would have to buy
`that it excludes this idea of this precomputed route.
` The other thing, if you go to slide 85, FatPipe's
`construction would require that the router updates have it
`faster than once per second.
` Now, patent owner argues that Karol doesn't have a
`packet path selector which selects between network interfaces
`on a per-packet basis because essentially what it is, your
`Honors, is that what's limited to OSPF, and OSPF is too slow.
` This dispute the fact that there is no minimum
`packet rate required anywhere in the patent. Mr. Williams,
`FatPipe's expert, confirmed that there is no minimum packet
`rate required by the patents.
` Under FatPipe's contention, by the way, Karol's
`
`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
`26
`
`14
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`routing data basis which are updated approximately once every
`second, that would mean that Karol meets the claim element
`when the packet rate is one packet per second or slower.
` But the fact is that Karol is not limited to OSPF,
`and we'll see this in several of the arguments today.
` Karol specifically says that a person of ordinary
`skill in the art, which is again why I mentioned at the
`outset that the level here is relatively high, but Karol
`specifically says that a person of ordinary skill in the art
`can implement faster protocols.
` Mr. Williams admits that Karol is not limited to
`OSPF, so there is no reason that -- to require into this
`claim that doesn't say anything about packet rate
`requirements that you would include one.
` So if the court has no other questions on per
`packet, I would like to turn to a related dispute over per
`session. We actually linked them together.
` JUDGE ZADO: Actually, I do have one question on
`per packet, and this goes back again to the specification of
`the 235 patent at column 14.
` It's the selection that Judge White had pointed out
`starting around line 44.
` MR. FITZPATRICK: Yes.
` JUDGE ZADO: That the selection can be performed
`once per packet or if they pertain to multiple packets, and
`I'm still struggling with the idea that the claim language
`
`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
`26
`
`15
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`does say specifically that there's a selection on a
`per-packet basis. The claim does not recite per-packet basis
`or per multiple packets, and so my question is that the way
`I'm reading this is it doesn't really look like the patent
`owner is attempting to read out a preferred embodiment, but
`rather there are two embodiments given in the specification,
`and the patent owner chose to claim one of those.
` And so if multiple packets were intended to be
`covered by this claim, wouldn't the claim have just said on a
`per-packet basis or per multiple packets?
` MR. FITZPATRICK: I think it's to the contrary,
`your Honor. If they wanted to say that every time a packet
`came in to the controller that packet was analyzed and that
`the decision tree was gone through, they could have said
`that.
` They didn't. And, in fact, when you look at,
`again, that same part of column 14, 44 through 46, I'm not
`sure how it couldn't read that out.
` There is certainly an embodiment here. It's clear
`here that this path selecting step, 908, may be performed
`once per packet, it could happen, or a given selection may
`pertain to multiple packets.
` So if claim 4 is read the way that FatPipe seeks to
`read it, you couldn't have that or a given selection may
`pertain to multiple packets because the selection for one
`would not apply to others.
`
`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
`26
`
`16
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
` And the same thing when you again look further down
`on column 16, 22 through 27, where -- when they're talking
`about load balancing and the reliability criteria where an
`initial path selection is made, and then the loop occurs in
`which multiple packets are received and then sent over the
`selected path without repeating the selecting step.
` If they wanted that selecting step to be done every
`time a packet came in, they would have claimed it
`differently. They certainly wouldn't have said what they
`said, and they certainly wouldn't have to then retrofit their
`construction.
` It certainly couldn't be the broadest reasonable
`interpretation to read out this embodiment that is now not
`only shown in figure 9 but also then described twice at both
`column 14 and column 16.
` JUDGE ZADO: So, if I understand correctly, it
`sounds like what you're saying is that the “or a given
`selection may pertain to multiple packets” is also part of the
`invention. The fact that you could do one or the other?
` MR. FITZPATRICK: I think that's certainly part of
`the embodiment. I think that's certainly part of what is
`certainly part of what's claimed, yes.
` JUDGE ZADO: And if the claim had read, rather than
`on a per-packet basis, if the claim had read “select between
`network interfaces once per packet?”
` MR. FITZPATRICK: I think that would be different.
`
`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
`26
`
`17
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`If it was once per packet, I think that would be a different
`situation, yes.
` JUDGE ZADO: Okay. Thank you. I had one other
`question because there was a lot of discussion so far about
`the -- I believe it was termed the high level skill of the
`art. Do you believe there is a dispute between the parties
`as to what the level of skill in the art is?
` MR. FITZPATRICK: So, the answer is really no, your
`Honor. I think the different -- we said that it's Talari's
`position that it is a Bachelor of Science in computer
`science, computer engineering, electrical engineering or
`equivalent field as well as at least two years of academic or
`industry experience in any type of networking field, and
`FatPipe stated that the experience would also include network
`design, configuration and/or diagnosis.
` I'm not sure exactly what those things add, but,
`no, I don't think the parties have a dispute into -- in fact,
`I think they agree that it's relatively similar and
`relatively high.
` JUDGE ZADO: Okay. Thank you.
` MR. FITZPATRICK: So a related -- if we could go to
`the next disputed issue, and it's related. Here this is the
`per-session dispute.
` And this is where FatPipe just ignores the explicit
`decision in Karol which shows a packet carrying the UDP
`segment as part of a session.
`
`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
`26
`
`18
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
` Now, FatPipe concedes, by the way, that Karol did
`discloses per session per claim 10. This is important.
`Instead of per-packet basis, what we just discussed, claims
`10 and claim 19 claim a per-session basis. And it's telling
`that FatPipe's only challenge to the per-session basis
`pertains to claim 19.
` For claim 10 it doesn't challenge that Karol's
`disclosure of a selecting step to make network path
`selections on a per-session basis isn't met.
` So claim 19 requires a packet path selector which
`selects between the network interfaces on a per-session
`basis, and then it adds to promote load balancing.
` Now, your Honors I'm sure saw that the 235 patent
`says very little about sessions. FatPipe really says nothing
`at all.
` What the 235 patent does disclose is that load
`balancing is preferably done on a per-packet basis for site
`to site data traffic or on a TCP or UDP session basis for
`Internet traffic and it goes to prior approaches of the per
`department, the per accounting, the per sales department or
`per router basis for dividing traffic.
` Now, if we turn to slide 72, we can see that Karol
`likewise discloses that certain packets carrying either TCP
`or UDP segments with certain sessions or applications, and it
`lists several, web access, Telemet and the like, are
`appropriate for a flow to the CO network and others for the
`
`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
`26
`
`19
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`CL network as shown in figure 6 of Karol.
` So in figure 6 what we see is step 635
`determines -- can we get that -- can your Honor see that?
` JUDGE WHITE: I have Karol up in front of me, so
`I'm looking at figure 6.
` MR. FITZPATRICK: Figure 6 of the step 635
`determines if the packet associated with UDP is associated
`with a session, and if yes, the CO network is selected in
`step 637.
` So there Karol selects between the network
`interface on a per-session basis.
` Now, obviously also should probably add packets
`carrying TCP segments within a session or application are
`routed to the CL network in figure 6 at step 625.
` But, you know, when you talk about per-session
`basis to promote load balancing, which is what the claim
`requires here, and if we can turn to slide 88, this is
`exactly what Karol discloses.
` Karol discloses these sessions or flows can be
`directed to one network based on bandwidth availability
`criterion, and when one path runs out of available bandwidth
`traffic, it can be put on the next best path. So that's
`Karol disclosing the load balancing associated with per
`session.
` This is also exactly how FatPipe read its claim
`when making its infringement allegation. FatPipe satisfied
`
`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
`26
`
`20
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`this claim element in the District Court litigation saying
`that traffic is put on the best path until that path runs out
`of available bandwidth. The additional traffic will then
`spill over to the next best path.
` So I think there's less of a dispute on per session
`than there was on per packet.
` The next area I would like to turn to is one where
`there is probably more dispute, and that is site interface.
`If we could turn to slide 90.
` The site interface issue applies to claims 4 and 9
`of the 235 patent and claims 1 and 13 of the 048 patent.
` Now, Karol disclose a site interface connecting a
`controller to a site. FatPipe really wants to ignore the
`description in Karol and look only at figure 1 of Karol.
` So, FatPipe is saying that the controller is a
`single device that has to be directly connected to the site.
` The claims certainly do not require the controller
`to be a single device directly connected to the site.
` Likewise, as we see in slide 90 and the 235 patent
`discloses, the controller can be and may be directly
`connected to the site, but it also can be connected to the
`site via a LAN.
` And interesting, again, FatPipe did not need a
`direct connection to satisfy its infringement allegations.
`This is exactly when FatPipe -- and that's what you see below
`when FatPipe served its infringement contentions, it
`
`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
`26
`
`21
`
`
`
`Case IPR2016-00976 (Patent 6,775,235 B2)
`Case IPR2016-00977 (Patent 7,406,048 B2)
`
`contradicted their argument to the Board.
` In there it said that if the controller is Talari's
`device, there is a LAN and a LAN switch between the Talari
`device and the site.
` But really besides the fact that the claims do not
`require that the controller be a single device directly
`connected to the site. Even if a direct connection is
`required, Karol discloses it.
` The express -- there is an expressed disclosure in
`Karol -- there are actually two that I would like to point
`out to the Board. One is Karol at column 2, lines 31 through
`34, where it states it is to be noted that the gateway of the
`present invention can be incorporated within a CL node. And
`in that sense, it would be directly connected.
` And, secondly, Karol also says at column 5, lines 5
`through 8, that the source or destination may be directly
`connected to the CL-CO gateway as opposed to being connected
`through a CL node.
` So this disclosure is referenced in the embodiment
`shown in figure 1 which we cited repeatedly in the position
`in the reply and in Dr. Ney's declaration, and, of course,
`figure 1 describes that embodiment where the controller
`really can be anywhere.
` And so I think that there is also this idea that
`the router source can create packets, and the CL node 111
`could also be the source.
`
`1
`2
`3
`4
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
