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
`
`__________________
`
`
`
`EXCLUSIVE LICENSEE FATPIPE INC.’S PRELIMINARY RESPONSE
`UNDER 37 C.F.R. § 42.107
`
`
`
`
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`TABLE OF CONTENTS
`
`Page
`EXHIBIT LIST .......................................................................................................... i
`I.
`INTRODUCTION .......................................................................................... 1
`PETITIONER’S REQUIREMENTS TO HAVE AN INTER PARTES
`II.
`REVIEW INSTITUTED ................................................................................ 2
`A. ANTICIPATION .................................................................................. 4
`B. OBVIOUSNESS .................................................................................. 4
`C.
`CLAIM CONSTRUCTION ................................................................. 8
`III. DESCRIPTION OF THE ’235 PATENT, THE PRIOR ART, AND
`THE PATENTABLY DISTINGUISHABLE FEATURES OF THE
`’235 PATENT ................................................................................................. 8
`A.
`THE ’235 PATENT.............................................................................. 8
`B. KAROL .............................................................................................. 13
`THE PETITION FAILS TO SHOW THAT THERE IS A
`C.
`REASONABLE LIKELIHOOD THAT THE PETITIONER
`WOULD PREVAIL WITH RESPECT TO ALL
`CHALLENGED CLAIMS IN GROUNDS 1-3 ................................. 23
`GROUND 1, CLAIM 11, 19 AND GROUNDS 2 AND
`1.
`3, CLAIMS 11-13, 19 - NO LOAD BALANCING
`ACROSS NETWORKS ........................................................... 25
`GROUNDS 1-3, CLAIM 19 - NO CLAIMED
`SECURITY FEATURES ......................................................... 34
`GROUNDS 1-3, CLAIM 5 - NO TWO ADDRESS
`RANGES .................................................................................. 38
`GROUNDS 1 AND 3, CLAIMS 4 AND 9 - NO PER
`PACKET BASIS ...................................................................... 45
`GROUNDS 1 AND 3, CLAIM 4 - NO PATH
`SELECTION FACTORS ......................................................... 51
`PETITIONER HAS FAILED TO PROVIDE A PROPER
`GRAHAM FACTOR ANALYSIS FOR GROUNDS 2 AND 3 ......... 53
`
`2.
`
`3.
`
`4.
`
`5.
`
`D.
`
`
`
`
`
`
`
`
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`TABLE OF CONTENTS
`
`Page
`
`1.
`
`2.
`
`GROUNDS 2 AND 3, CLAIM 5 - NO GRAHAM
`FACTOR .................................................................................. 54
`GROUND 3, CLAIMS 4 AND 9 - NO GRAHAM
`FACTOR .................................................................................. 54
`GROUNDS 1-3 ARE CUMULATIVE ................................... 55
`3.
`IV. THE PETITION FAILS TO PUT FORTH GROUNDS FOR WHICH
`RELIEF SHOULD BE GRANTED ............................................................. 56
`A. GROUND 3 IS FACIALLY DEFECTIVE BECAUSE IT
`RELIES ON ALTERNATIVE CONSTRUCTIONS ........................ 56
`IMPROPER INCORPORATION BY REFERENCE TO THE
`NEGUS DECLARATION AND UNCITED EXHIBITS ................. 60
`CONCLUSION ............................................................................................. 61
`
`B.
`
`V.
`
`
`
`
`
`
`
`
`
`
`
`
`

`
`
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`EXHIBIT LIST
`
`Exhibit 2001
`
`
`
`
`Declaration of Joel Williams in Support of Preliminary
`Response
`
`
`
`i
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`INTRODUCTION
`Talari Networks, Inc. (“Petitioner”) filed a Petition for Inter Partes Review
`
`I.
`
`
`of U.S. Patent No. 6,775,235 B2 (the “’235 Patent”) on April 29, 2016 (Paper 1, the
`
`“Petition”). The Patent Trial and Appeal Board (the “Board”) mailed a Notice of
`
`5
`
`Filing Date Accorded to Petition on May 3, 2016 (Paper 3). Pursuant to 37 C.F.R.
`
`§ 42.107, exclusive licensee FatPipe, Inc. (for the purposes of consistency with
`
`Board convention, FatPipe will be referred to as “Patent Owner”) timely submits
`
`this Preliminary Response.
`
`Patent Owner respectfully requests the Board to deny the Petition because it
`
`10
`
`does not satisfy the statutory requirements of 35 U.S.C. § 311 and 37 C.F.R. §
`
`42.100. In particular, Petitioner’s request for inter partes review should be denied
`
`for at least the following reasons:
`
`1. With respect to all challenged claims of Grounds 1-3, the Petition fails to
`
`illustrate that the cited prior art teaches several claim elements and fails to
`
`15
`
`establish that there is a “reasonable likelihood that the petitioner would
`
`prevail” as required by 35 U.S.C. § 314.
`
`2. In addition, with respect to Grounds 2 and 3, the Petition fails to describe
`
`the scope and content of the prior art, and the differences between the
`
`claimed subject matter and the asserted references. See Graham v. John
`
`20
`
`Deere Co., 383 U.S. 1, 17–18 (1966).
`
`
`
`-1-
`
`

`
`3. In addition, Ground 3 is based entirely on alternative constructions. The
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Petition fails to describe “[h]ow the challenged claim is to be construed” as
`
`required by 37 C.F.R. § 42.104(b)(3).
`
`4. Finally, the Petition contains conclusory statements and impermissible
`
`5
`
`incorporation by reference to arguments in the Negus Declaration in
`
`violation of 37 C.F.R. § 42.6(a)(3).
`
`Each of these reasons will be discussed in further detail below after a brief
`
`description of a Petitioner’s requirements to have its Petition instituted.
`
`II.
`PETITIONER’S REQUIREMENTS TO HAVE AN INTER PARTES
`REVIEW INSTITUTED
`
`10
`
`
`
`35 U.S.C. § 314(a) sets forth the threshold requirements for instituting an
`
`inter partes review:
`
`(a)THRESHOLD.—The Director may not authorize an inter partes review
`
`to be instituted unless the Director determines that the information
`
`15
`
`presented in the petition filed under section 311 and any response filed
`
`under section 313 shows that there is a reasonable likelihood that the
`
`petitioner would prevail with respect to at least 1 of the claims challenged
`
`in the petition.
`
`(Emphasis added). The Petition must contain a “full statement of the reasons for
`
`20
`
`the relief requested, including a detailed explanation of the significance of the
`
`
`
`-2-
`
`

`
`evidence including material facts, and the governing law, rules, and precedent.”
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`37 C.F.R. § 42.22(a)(2). To establish a reasonable likelihood, the Petition “must
`
`specify where each element of the claim is found in the prior art patents or printed
`
`publications relied upon.” See 37 C.F.R. § 42.104(b)(4). The Petition “must
`
`5
`
`identify . . . [h]ow the challenged claim is to be construed.” C.F.R. §
`
`42.104(b)(3).
`
`
`
`Petitioner bears the “burden of proof to establish that it is entitled to the
`
`requested relief.” 37 C.F.R. § 42.20(c). “[O]nly the basis, rationale, and
`
`reasoning put forth by the Petitioner in the Petition” should be addressed and “all
`
`10
`
`vagueness and ambiguity in Petitioner’s arguments [should be resolved] against
`
`the Petitioner.” See Whole Space Industries Ltd. v. Zipshade Industrial (B.V.I.)
`
`Corp., IPR2015-00488, Paper 14 at 8 (July 24, 2015); See also Liberty Mutual Ins.
`
`Co. v. Progressive Casualty Ins. Co., Case CBM-2012-00003, slip op. at 10
`
`(PTAB Oct. 25, 2012) (Paper 8).
`
`15
`
`
`
`“Arguments must not be incorporated by reference from one document into
`
`another document.” 37 C.F.R. §42.6(a)(3). No relief should be granted unless the
`
`arguments in the Petition itself support granting such relief. See Whole Space,
`
`IPR2015-00488 at 8-9; see also Hopkins ManufacturingCorp. v. Cequent
`
`Performance Products, Inc., IPR2015-00616, Paper 9 at 7 (August 17, 2015);
`
`20
`
`Cisco Systems, Inc., v. C-Cation Technologies, LLC, IPR2014-00454, Paper 12 at
`
`
`
`-3-
`
`

`
`10 (August 29, 2014); see also Plant Science, Inc. v. The Andersons, Inc.,
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`IPR2014-00939, Paper 8 at 15 (December 14, 2014) (“This practice of citing the
`
`Sensibaugh Declaration to support conclusory statements that are not otherwise
`
`supported in the Petition amounts to incorporation by reference.”).
`
`5
`
`
`
`A. ANTICIPATION
`“It is axiomatic that for anticipation, each and every claim limitation must be
`
`explicitly or inherently disclosed in the prior art.” In re NTP, INC., 654 F.3d 1279,
`
`1302 (Fed. Cir. 2011) (citations omitted). “The fact that a certain result or
`
`characteristic may occur or be present in the prior art is not sufficient to establish
`
`10
`
`the inherency of that result or characteristic.” See MPEP § 2112 (citing In re
`
`Rijckaert, 9 F.3d 1531, 1534, 28 USPQ2d 1955, 1957 (Fed. Cir. 1993)). The
`
`“allegedly inherent characteristic [must] necessarily flow[] from the teachings of
`
`the applied prior art.” Id. (citing Ex parte Levy, 17 USPQ2d 1461, 1464 (Bd. Pat.
`
`App. & Inter. 1990)).
`
`15
`
`20
`
`
`
`
`
`
`B. OBVIOUSNESS
`As explained by the Federal Circuit:
`
`Obviousness requires more than a mere showing that the prior art
`includes separate references covering each separate limitation in a
`claim. . . . Rather, obviousness requires the additional showing that a
`person of ordinary skill at the time of the invention would have
`
`-4-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`selected and combined those prior art elements in the normal course
`of research and development to yield the claimed invention.
`
`(Unigene Labs., Inc. v. Apotex, Inc., 655 F.3d 1352, 1360 (Fed. Cir. 2011)
`
`5
`
`(citations omitted).) “To render a claim obvious, prior art cannot be ‘vague’ and
`
`must collectively, although not explicitly, guide an artisan of ordinary skill towards
`
`a particular solution. . . . [A] combination is only obvious to try if a person of
`
`ordinary skill has ‘a good reason to pursue the known options.’” Id. at 1361
`
`(quoting KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007)). “[A]ny need or
`
`10
`
`problem known in the field of endeavor at the time of invention and addressed by
`
`the patent can provide a reason for combining the elements in the manner
`
`claimed.” KSR, 550 U.S. at 420. Further, “if a technique has been used to improve
`
`one device, and a person of ordinary skill in the art would recognize that it would
`
`improve similar devices in the same way, using the technique is obvious unless its
`
`15
`
`actual application is beyond that person’s skill.” Id. at 417; but see Abbott Labs v.
`
`Sandoz, Inc., 544 F.3d 1341, 1352 (Fed. Cir. 2008) (“[K]nowledge of the goal does
`
`not render its achievement obvious.”).)
`
`In proposing that a POSITA at the time of the invention would have
`
`combined the references in a particular way to meet the claimed invention, an
`
`20
`
`obviousness analysis must support the proposed combination with “some
`
`articulated reasoning with some rational underpinning.” KSR, 550 U.S. at 418
`
`
`
`-5-
`
`

`
`(quoting In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006)). A proposed
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`combination cannot be supported based on “merely conclusory statements.” Kahn,
`
`441 F.3d at 988. The “articulated reasoning with rational underpinning” must be
`
`found in the petition itself. Whole Space, IPR2015-00488, Paper 14 at 8.
`
`5
`
`However, “[w]hen the prior art teaches away from combining certain known
`
`elements, discovery of successful means of combining them is more likely to be
`
`nonobvious.” KSR, 550 U.S. at 416.
`
`The question of obviousness is resolved on the basis of underlying factual
`
`determinations, including: (1) the scope and content of the prior art; (2) any
`
`10
`
`differences between the claimed subject matter and the prior art; (3) the level of
`
`skill in the art; and (4) where in evidence, so-called secondary considerations. See
`
`Graham, 383 U.S. 1, 17–18. “[C]ourts must consider all of the Graham factors
`
`prior to reaching a conclusion with respect to obviousness.” Kinetic Concepts, Inc.
`
`v. Smith & Nephew, Inc., 688 F.3d 1342, 1360 (Fed. Cir. 2012); see also Whole
`
`15
`
`Space, IPR2015-00488, Paper 14 at 9 (“[N]or does the Petition identify the
`
`differences between the prior art and the claims at issue, which is one of the
`
`fundamental factual inquiries underlying an obviousness analysis.”); Google, Inc.
`
`v. Everymd.com LLC, IPR2014-00347, Paper 9 at 25 (May 22, 2014) (“Rather,
`
`Petitioners’ summaries, quotations, and citations from both references, with
`
`20
`
`Belanger’s figures, place the burden on us to…identify any differences between
`
`
`
`-6-
`
`

`
`the claimed subject matter and the teachings of Shah and Belanger.”); Liberty
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Mutual, CBM2012-00003, Paper 7 at 2 -3 (“Differences between the claimed
`
`invention and the prior art are a critically important underlying factual inquiry for
`
`any obviousness analysis.”).
`
`5
`
`If a petitioner “does not state the differences between a challenged claim and
`
`the prior art, and relies instead on the Patent Owner and the Board to determine
`
`those differences based on the rest of the submission in the petition,” then such a
`
`petitioner “risks having the corresponding ground of obviousness not included for
`
`trial for failing to adequately state a claim for relief.” Liberty Mutual, CBM2012-
`
`10
`
`000033, Paper 7 at 3. The scope and content of the prior art and the differences
`
`between the prior art and the claims must be explained. Id. at 3 (“The first
`
`[deficiency] involves a plurality of prior art references applied not in combination
`
`to complement one another but as distinct and separate alternatives.”). “Vague
`
`statements or hints of differences [in obviousness combinations] not only burdens
`
`15
`
`the Board, but puts a patent owner somewhat of a disadvantage with having to
`
`guess what any differences a petitioner may believe exi[s]t.” See Volkswagen Grp.
`
`of America, Inc. v. Emerachem Holdings, IPR2014-01556, Paper 19, p. 17 (PTAB
`
`2015).
`
`
`
`-7-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`C. CLAIM CONSTRUCTION
`In an inter partes review, claim terms in an unexpired patent are interpreted
`
`according to their “broadest reasonable construction in light of the specification of
`
`the patent” in which they appear. 37 C.F.R. § 42.100(b); see Cuozzo Speed Techs.,
`
`5
`
`LLC v. Lee, 136 S. Ct. 2131, 2144 (2016) (holding that 37 C.F.R. § 42.100(b)
`
`“represents a reasonable exercise of the rulemaking authority that Congress
`
`delegated to the . . . Office”). Claim terms should be interpreted according to their
`
`ordinary and customary meaning in the context of the patent’s written description.
`
`See In re Translogic Tech., Inc., 504 F.3d 1249, 1257 (Fed. Cir. 2007). At this
`
`10
`
`preliminary stage, Patent Owner has not offered a construction but reserves the
`
`right to offer constructions should trial be instituted on any claim.
`
`III. DESCRIPTION OF THE ’235 PATENT, THE PRIOR ART, AND THE
`PATENTABLY DISTINGUISHABLE FEATURES OF THE ’235 PATENT
`A. THE ’235 PATENT
`The ’235 patent is directed to “two or more disparate networks in parallel”
`
`15
`
`that “provide load balancing across network connections, greater reliability, and/or
`
`increased security.” ’235 patent, Abstract; Ex. 2001, ¶37. When “all attached
`
`disparate networks are operating, one controller preferably balances the load
`
`between them.” Id.
`
`
`
`-8-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Prior art approaches did not combine two or more disparate networks in
`
`parallel to provide benefits such as dynamic per-packet load-balancing. See ’235
`
`patent, 4:39-45; Ex. 2001, ¶37. Prior art Fig. 2 (below), for example, used a
`
`primary network (the frame relay network 106) and only used the secondary
`
`5
`
`network (the ISDN network 204) when the primary network failed. See ’235
`
`patent, 3:18-28. The primary network path is used for most or all of traffic while
`
`the other path is used only when the primary path fails. Id. at 9:55-65. The prior
`
`art configuration of Fig. 2 does not consider load balancing on a packet-by-packet
`
`basis, or provide security by splitting pieces of given messages between disparate
`
`10
`
`networks. Id. at 9:65-10:3.
`
`
`
`Other approaches did not provide dynamic load balancing amongst multiple
`
`disparate networks. ’235 patent, 2:56-65; Ex. 2001, ¶38. Prior art Fig. 1 (below),
`
`for example, requires that networks agree upon factors relating to communications
`
`
`
`-9-
`
`

`
`prior to traffic being sent. Id. at 2:52-55. They can provide a rough balance by
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`sending different types of traffic or flows through particular routers (e.g., router A
`
`or router B), but this does not balance router loads dynamically in response to
`
`actual traffic. Id. at 2:56-65. The prior art approaches required set-up with broad
`
`5
`
`granularity and did not load-balance dynamically in response to actual traffic. Id.
`
`at 9:4-9. Other prior art approaches, such as those described in Figs. 3-4, did not
`
`provide networks in parallel (3:29-4:4) and could not provide load-balancing or
`
`improve reliability. Id. at 3:63-4:4.
`
`
`
`10
`
`Other parallel networks, such as those in Fig. 5, did not have the “fine
`
`grained packet routing of the present invention.” ’235 patent, 5:24-28; Ex. 2001,
`
`¶39. These networks only had coarse routing of traffic or flows where “all packets
`
`from department X might be sent over the frame relay connection 106 while all
`
`packets from department Y are sent over the Internet 500. Or the architecture
`
`15
`
`might send all traffic over the frame relay network unless that network fails. . .”
`
`
`
`-10-
`
`

`
`Id. at 4:18-22. These described prior art architectures did not provide dynamic
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`packet-by-packet routing between disparate networks. Ex. 2001, ¶39.
`
`
`
`The ’235 patent describes numerous parallel networks that can be of many
`
`5
`
`different types (7:6-20) where the networks are divided and routed by known
`
`address ranges. ’235 patent, 8:23-28; Ex. 2001, ¶40. These ranges, for example,
`
`can include 192.168.x.x for a LAN, 200.x.x.x for the Internet, or 196.x.x.x for a
`
`Frame Relay. ’235 patent, 8:50-53. Packets can be re-routed to different networks
`
`by changing their destination address ranges. Id. at 9:12-29, 13:39-57. For
`
`10
`
`example, a packet bearing a destination address 10.0.x.x can be changed to 198.x.x
`
`x to route it through the frame relay network. Id. at 9:12-29. This provided for
`
`easy routing of packets between disparate networks. “Without the invention, . . .
`
`network devices are pre-configured . . . such that all such packets with [a given]
`
`destination address must be sent to [the addressed network], even though there is
`
`15
`
`[second network] connectivity between the two locations.” Id. at 8:55-63.
`
`
`
`-11-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`The ’235 patent also describes routing packets to improve reliability,
`
`security, and to balance loads in parallel networks. ’235 patent, 4:39-45; Fig. 9;
`
`13:33-38; Ex. 2001, ¶41. As illustrated in the flowchart of Fig. 9 (below), the path
`
`selection may use load balancing, connectivity, or security criterion. Id. at 14:59-
`
`5
`
`15:3; 11:11-11:63. Loads can be balanced across the parallel disparate networks,
`
`on a per-packet basis, to load-balance after packets leave a network interface. Id.
`
`at 11:24-27. Packet-by-packet load balancing provides a finer granularity than
`
`what was found in the prior art. Id. at 9:12-19. Additionally, to improve security,
`
`messages can be divided up between networks. Id. at 11:46-49. These criteria are
`
`10
`
`applied on parallel networks, not those which use one network as a fail-over or as
`
`an alternative for another network. Id. at 9:55-65.
`
`
`
`
`
`-12-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`B. KAROL
`Karol is directed to routing traffic flows to a connection oriented (CO)
`
`network from a connectionless (CL) network. Karol, Abstract; Ex. 2001, ¶42.
`
`Karol states that a CL network has traffic that can be routed onto a connection for a
`
`5
`
`CO network. Karol, Abstract. The CO network can be an ATM (Asynchronous
`
`Transfer Mode) or WDM (Wavelength Division Multiplexed) network, for
`
`example, and the CL network can be an IP protocol network. Id. at 1:8-17; 2:53-
`
`63. CO connections need to be set-up before use. Id. at 1:24-39 (stating that
`
`switched mode is set up prior to information transfer and provisioned mode is set
`
`10
`
`up when the network is provisioned). But, once set up, CO connections are used
`
`for flows whose service contracts authorize CO service. See id. at Fig. 5, 503;
`
`5:35-46.
`
`Traffic flows between the CO and CL networks are controlled by nodes
`
`called CL-CO gateways. See id. at 2:13-19; Ex. 2001, ¶43. CL-CO gateways
`
`15
`
`handle (1) packets when the CO network connections are being set up; (2) routing
`
`flows from the CL network to the CO network; and (3) flows whose service
`
`contracts do not indicate CO network service. Karol, 7:64-8:2. The CL-CO
`
`gateways also set-up CO network connections for the CO paths, and determine if
`
`packets of those flows should be halted or buffered (id. at 12:23-37) while the
`
`20
`
`connections are being set up. Id. at 1:24-28; 9:46-48; 10:8-15; 10:45-47; 10:61-63.
`
`
`
`-13-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Fig. 1, for example, illustrates a “parallel architecture” that illustrates “how
`
`to handle traffic . . . arriving at the CL-CO gateway until the desired connection is
`
`established in the CO network.” Karol, 4:12-16; Ex. 2001, ¶44. As illustrated
`
`below in Fig. 1, traffic from source 101 arrives from a CL network 110 to a CL-
`
`5
`
`CO gateway 140. Karol, 4:36-67. Whether the CL-CO gateway 140 uses the CO
`
`network depends on user-specified service requirements that have been pre-defined
`
`and set up before-hand. Id. at 5:35-38. Whether to use either the CL or the CO
`
`network for particular flows is based on user-specified service requirements. Id. at
`
`5:34-57; 15:20-31; 16:3-8.
`
`10
`
`
`
`If the CO network is set to be used by service requirements, but a connection
`
`is yet to be set-up, packets are turned around onto the CL network (2:41-45; claims
`
`11, 15), halted, slowed-down until the CO network is setup (4:16-29) or buffered
`
`(7:1-13). Ex. 2001, ¶45. Packets that are turned around are source routed until the
`
`
`
`-14-
`
`

`
`CO connection is set up. Karol, 4:16-18. The CL-CO gateway “turn[s] back IP
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`datagrams to the CL network using IP source routing to override routing tables at
`
`the routers.” Id. at 8:51-57. Karol states that “source routing is used to force
`
`intermediate routers to use the source route carried in the datagram header instead
`
`5
`
`of the path indicated by their precomputed routing tables.” Id. at 11:27-31. Source
`
`routing involves changing the packet to add routing instructions to prevent loops.
`
`See id. A POSITA understands that this is typically accomplished by replacing the
`
`original packet with a “Source Demand Route Packet” which contains a new IP
`
`Header and the original data payload. Ex. 2001, ¶46.
`
`10
`
`The particular routing of flows to the CO or CL network in the CL-CO
`
`gateway is operated by “a processor 430 and associated database 431 for
`
`controlling the gateway packet handling operations and for storing forwarding,
`
`flow control, header translation and other information.” Karol, 6:40-44; Ex. 2001,
`
`¶47. The database 431 includes “a series of individual databases arranged to store
`
`15
`
`information used in various of the functions performed by processor 430, and may
`
`include, as an example, a datagram forwarding database 432, a flow database 433,
`
`and a header translation database 434.” Karol, 7:36-40. The datagram forwarding
`
`database 432 “stores the next hop router address and outgoing port number
`
`corresponding to each destination.” Id. at 7:36-41. The flow database 433 “stores
`
`20
`
`information used to determine how to handle packets from flows requiring a
`
`
`
`-15-
`
`

`
`connection oriented service.” Id. at 7:42-44. The header translation database 434
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`“indicates the incoming CL packet header field values and the corresponding CO
`
`packet header field values.” Id. at 7:55-59. The database 431 (which contains the
`
`forwarding database 432, flow database 433, and translation database 434) is not
`
`5
`
`dynamic; it handles particular flows and how to route them to the CO network, if
`
`the CO network is dictated by service contract. Id. at 5:34-57; 15:20-31; 16:3-8.
`
`Further, protocol converter 450 “generates CO packets” derived from the CL
`
`packets. Id. at 2:19
`
`
`
`10
`
`
`
`As shown in the Fig. 5 excerpt (below), when an IP datagram arrives at the
`
`CL-CO gateway of Fig. 4, step 503 first determines “whether the flow should be
`
`handled via the CO network or not” based on whether the “packet [is] from a flow
`
`that needs CO service.” Karol, 8:61-62; Fig. 5, 503; Ex. 2001, ¶48. The
`
`determination as to whether to use a CO network is defined by user-specified
`
`
`
`-16-
`
`

`
`service requirements well in advance. Karol, 5:34-57; 15:20-31; 16:3-8. If a user
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`does not specify that a CO network is used, packets are turned back to the CL node
`
`using source-routing or dropped. 9:26-36. If a CO network is specified to be used,
`
`the packet is sent to the packet buffer 440 for usage in the CO network. Id. 8:63-
`
`5
`
`9:1.
`
`
`
`
`
`Figs. 5-7 describe a system where, if a CO flow is specified by a user’s
`
`service contract, it is used; if it is not specified by user contract, it does not use the
`
`CO network. Karol, 3:6-16; 5:35-38; 15:20-31; 16:3-8; Ex. 2001, ¶49. Fig. 6 also
`
`10
`
`illustrates steps for determining if packets are TCP or UDP. Karol, 9:37-43. For
`
`UDP, once the CO is available and specified by user contract, it is used. Id. at
`
`12:1-5; 5:35-37. The other steps are related to opening a session on the CO
`
`network: (step 631) the system determines whether the application has an end-to-
`
`end handshake (i.e., a CO connection) prior to data transfer; (step 635) if the UDP
`
`
`
`-17-
`
`

`
`packet is related to opening a session; or (step 625) if the UDP packet should be
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`forwarded to the CL network. Id. at 10:52-11:8.
`
`
`
`
`
`TCP packets use the CO network if it is available and specified by service
`
`5
`
`contract. Karol, 8:61-65; 5:35-37; Ex. 2001, ¶50. The other steps are related to
`
`opening or closing a session on the CO network: TCP SYN packets request a CO
`
`connection setup (9:46-48), TCP FIN or RST packets close a particular CO flow
`
`(11:8-13; 10:24-28; 9:42:51; 11:8-13), and TCP data segments use the CO network
`
`if it is available and dictated by service contract. Karol, 8:33-41; 5:23-38; 11:8-13;
`
`10
`
`15:20-31; 16:3-8.
`
`
`
`Figs. 9-11 of Karol describe keeping routing databases (tables) consistent
`
`between gateways using the Open Shortest Path First (OSPF) routing protocol. Id.
`
`at 14:23-36; Ex. 2001, ¶51. OSPF allows CL-CO gateways to generate Link State
`
`Advertisements (LSAs) that “report point-to-point links” between themselves.”
`
`15
`
`Karol, 14:23-27. “[L]inks reported in the LSAs appear in the OSPF topology
`
`database of all the routers in the same OSPF area and have associated link
`
`
`
`-18-
`
`

`
`weights.” Id. at 14:30-32. A POSITA understands that a network that uses OSPF
`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`and LSAs operates too slowly to provide dynamic adjustments on a packet-by-
`
`packet basis because LSA weights need to be periodically transmitted across the
`
`network to update link-weights. See Ex. 2001, ¶52.
`
`5
`
`In Figs. 9-11, which utilize link weights, Karol describes the CO network as
`
`the fastest and preferred route if routing tables are used - the CO network is always
`
`shown as 1-hop. Karol, 14:24-49; 3:19-23 (stating that Fig. 10 is a link-weight
`
`representation of Fig. 9); Ex. 2001, ¶53. Routing tables can also be set up
`
`according to a user’s service contract. Karol, 15:20-31. Routing tables take
`
`10
`
`advantage of shorter paths (5:39-43), where links between nodes of the CL
`
`network may be weighted. Id. at 11:27-47. But, from the perspective of each CL-
`
`CO gateway that interfaces with both the CO and CL networks (e.g., gateways
`
`960-962 of Fig. 10), the 1-hop route (i.e., the CO network), is the fastest (i.e., 1-
`
`hop) path. Id. at 14:24-49; 3:19-23; Fig. 10 (Red Annotations, Below). Karol
`
`15
`
`consistently describes the CO network as the preferred network for data traffic if it
`
`is available. id. at 12:1-5.
`
`
`
`-19-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`
`Karol describes link-weighting only in context of CL network nodes, not CO
`
`
`
`nodes which, as described above, are always 1. Ex. 2001, ¶54. Karol states that
`
`links to CL-CO gateways from CL nodes may be weighted because CL-CO
`
`5
`
`gateways “have call handling capacities” and “weights associated with [those]
`
`links in the routing protocol can be dynamically adjusted. . .” Karol, 17:64-65. To
`
`ease pressure on the CL-CO gateways, “link weights [are adjusted]. . . to divert
`
`connections away from overloaded call processors.” Id. at 17:64-65; see also Fig.
`
`10 (Green Annotations, below). The weighted links are, however, from a
`
`10
`
`particular CL node to the CL-CO gateways – they are not weighted links to CO
`
`nodes. Id. at 17:44-62. This weighting does not affect the link-weight of the CO
`
`network from the perspective of the CL-CO gateways – the CO connections are
`
`always seen as 1-hop and preferred, regardless of any link-weighting on the CL
`
`network. Id. at 14:24-49; see also 3:19-23; see also Fig. 10 (Red Annotations,
`
`15
`
`below).
`
`
`
`-20-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Only links to GW1, GW2, and
`GW3 would be weighted
`
`From the perspective of the CL-CO
`gateways, the CO network is
`always fast – a 1-hop
`
`
`
`The architectures of Figs. 9-11, which illustrate link weighting, are also
`
`restrictive in choice of CO or CL networks. Ex. 2001, ¶55. As shown in annotated
`
`Fig. 11 below, from the perspective of the CL-CO gateway, a particular network
`
`5
`
`must be used based on the addressed endpoint. Id. For example, a packet at the
`
`CL-CO gateway (annotated in Blue, below) that is addressed for the Endpoint 1102
`
`must use the CO network (annotated in Red, below). Id. Similarly, a packet at the
`
`CL-CO gateway (annotated in Blue, below) that is addressed for any of Endpoints
`
`1101, 1103, 1104 must use the CL network. Id. For a packet at the CL-CO
`
`10
`
`gateway, there is no choice to use either one network or the other – one must
`
`always be used. Id.
`
`
`
`-21-
`
`

`
`Case IPR2016-00976, Patent 6,775,235 B2
`Patent Owner Preliminary Response
`
`
`Packets from CL-CO gateway
`to Endpoint 102 must go
`through the CO network
`
`Packets from CL-CO gateway
`to Endpoint 102 must go
`through the CO network
`
`
`
`Figs. 9-11 of Karol are in contrast to Fig. 10 of the ’235 patent, which allows
`
`Controller A (Blue, below) to choose, on a packet-by-packet basis, between the
`
`Internet 500 (Green, below) or the Frame Relay Network 106 (Red, below) to route
`
`5
`
`packets to Site B 102. ’235 patent at 9:12-29; Ex. 2001, ¶56. The ’235 patent
`
`states that routing through one network or another can be done on a packet-by-
`
`packet basis based upon certain criteria such as load balancing. Id. at 9:12-1