`571-272-7822
`
`Paper 77
`Entered: November 14, 2018
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`INTEL CORPORATION, CAVIUM, LLC, and DELL, INC,
`Petitioner,
`
`V.
`
`ALACRITECH, INC.,
`Patent Owner.
`
`Case IPR2017-01410
`
`Patent 8,131,880 B21
`
`Before STEPHEN C. SIU, DANIEL N. FISHMAN, and
`CHARLES J. BOUDREAU, Administrative Patent Judges.
`
`SIU, Administrative Patent Judge.
`
`FINAL WRITTEN DECISION
`
`35 US. C. § 318(a)
`
`1 Cavium, Inc., which filed a Petition in Case IPR2017—01737, and Dell,
`Inc., which filed a Petition in Case IPR2018-OO339, were joined as
`petitioners in this proceeding; According to updated mandatory notices filed
`in this proceeding, Cavium, Inc. has now been converted to Cavium, LLC.
`Paper 72.
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`1. INTRODUCTION
`
`Responsive to the filed Petition (Paper 1, “Pet.”), we instituted an
`
`inter partes review of all challenged claims (claims 32, 34, 35, 37—39, and
`
`41—43) ofU.S. Patent No. 8,131,880 B2 (“the ’880 patent,” Ex. 1001).
`
`Paper 8. Alacritech, Inc. (“Patent Owner”) filed a Corrected Patent Owner’s
`
`Response (Paper 32, “PO Resp”) and Intel Corporation filed a Petitioner
`
`Reply (Paper 42, “Pet. Reply”). Responsive to petitions and requests for
`
`joinder filed in IPR2017-01737 and IPR2018-00339, we joined Cavium, Inc.
`
`(now Cavium, LLC) and Dell, Inc., respectively, as petitioners in this
`
`proceeding. Paper 8 in IPR2017-01737; Paper 9 in IPR2018-003 39.
`
`According to updated mandatory notices filed in this proceeding, Cavium,
`
`Inc. has now been converted to Cavium, LLC. Paper 72. Petitioners Intel
`
`Corporation, Cavium, Inc, and Dell, Inc. are identified herein collectively as
`
`“Petitioner.”
`
`Patent Owner filed a Contingent Motion to Amend (Paper 20),
`
`Petitioner filed an Opposition to Patent Owner’s Contingent Motion to
`
`Amend (Paper 38), Patent Owner filed a Reply to Petitioner’s Opposition
`
`(Paper 43), and, pursuant to our having granted leave, Petitioner filed a Sur-
`
`Reply (Paper 50).
`
`Petitioner filed a Motion to Exclude (Paper 54), Patent Owner filed an
`
`Opposition (Paper 58), and Petitioner filed a Reply to Patent Owner’s
`Opposition (Paper 60).
`3
`
`Patent Owner filed a Motion to Exclude (Paper 55), Petitioner filed an
`
`Opposition (Paper 57), and Patent Owner filed a Reply to Patent Owner’s
`
`Opposition (Paper 61).
`
`'
`
`Patent Owner filed a Motion to Seal (Paper 30).
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`A transcript of an oral hearing held on September 13, 2018 (Paper 73)
`
`has been entered into the record.
`
`We have jurisdiction under 35 U.S.C. § 6. This Final Written
`
`Decision is issued pursuant to 35 U.S.C. § 318(a). We base our decision on
`
`the preponderance ofthe evidence. 35 U.S.C. § 316(c); 37 CPR. § 42.1(d).
`
`Having reviewed the arguments of the parties and the supporting
`
`evidence, we conclude that Petitioner has demonstrated by a preponderance
`
`of the evidence that the challenged claims are unpatentable. We also deny
`
`Petitioner’s Motion to Exclude, Dismiss Patent Owner’s Motion to Exclude,
`
`grant Patent Owner’s Motion to Seal, and deny Patent Owner’s Contingent
`
`Motion to Amend.
`
`THE ’880 PATENT (EXHIBIT 1001)
`
`The ’880 patent describes a system and method for performing
`
`network processing tasks on a network interface card. Ex. 1001, 3:45—47.
`
`ILLUSTRATIVE CLAIM
`
`Claim 32, reproduced below, is illustrative of the claimed subject
`
`matter:
`
`A method of transferring a packet received at a
`32.
`network interface from a network to a host computer system,
`
`comprising:
`receiving a packet from a network at a network interface of
`a host computer system;
`parsing a header portion of said packet to extract an
`identifier of a source entity and an identifier of a destination
`entity;
`‘
`generating a flow key from said source identifier and said
`destination identifier to identify a communication flow
`comprising said packet, wherein said flow key includes a TCP
`
`3
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`connection for the communication flow and a first hop medium
`access control (MAC) layer address;
`determining whether a header in said header portion
`conforms to a pre—selected protocol;
`storing said flow key in a database;
`associating an operation code with said packet, wherein
`said operation code identifies a status of said packet;
`storing said packet in a packet memory;
`if said header conforms to the TCP protocol:
`storing a data portion of said packet in a re-asSembly
`buffer;
`storing said header portion in a header buffer; and
`processing, by the network interface, said packet according
`to the TCP connection.
`
`Id. at 93:3~29.
`
`GROUNDS OF INSTITUTION
`
`We instituted trial on claims 32, 34, 35, 39, and 41—43 as unpatentable
`
`under 35 U.S.C. § 103 over Thia2 and Tanenbaum3 and claims 37 and 38 as
`
`unpatentable under 35 U.S.C. § 103 over Thia, Tanenbaum, and Nahum,4
`
`which are all proposed challenges to patentability. Pet. 19.
`
`ANALYSIS
`
`Petitioner argues that the combination of Thia and Tanenbaum
`
`discloses receiving a packet at a network interface of a host computer system
`
`2 Y.H. Thia and CM. Woodside, “A Reduced Operation Protocol Engine
`(ROPE) for a Multiple-Layer Bypass Architecture,” 1995 (“Thia,” Ex.
`1015)
`3 Andrew S Tanenbaum, Computer Networks, Third Edition, 1996
`(“Tanenbaum, ” Ex. 1006).
`4 Erich M Nahum, et a1., Performance Issuesin Parallelized Network
`' Protocols,” Proceedings of the First Symposium on Operating Systems
`Design and Implementation, November 1994 (“Nahum,” Ex. 1079).
`
`4
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`and parsing a header portion of the packet to extract an identifier of a source
`
`entity and an identifier of a destination entity as recited in claim 32. For
`
`example, Petitioner argues that Tanenbaum discloses that one of skill in the
`
`art would have understood that “conventional processing of an incoming
`
`data packet” includes receiving an “incoming TCP/IP packet [that]
`
`comprises both an IP header and a TCP header,” that “[t]he IP and TCP
`
`headers include source and destination IP addresses and TCP ports,” and that
`“the source and destination addresses and ports are identifiers of source and
`
`‘
`
`destination entities.” Pet. 44—46 (citing Ex. 1006, .431, .434, .524, .541,
`
`.544, Figs. 5-45, 6-23, 6-24).
`
`Petitioner argues that the combination of Thia and Tanenbaum
`
`discloses generating a flow key that includes a TCP connection for a
`communication flow and a first hop medium access control (MAC) layer
`
`address, as recited in claim 32. Pet. 47—49. For example, as Petitioner
`
`explains, Tanenbaum discloses a “connection record” or flow key that “is
`generated from ‘some simple function of the data .
`.
`. i.e., the two IP
`
`address and two port numbers” that “identif[ies] a TCP connection for the
`
`communication flow.” Pet. 48 (citing Ex. 1006, .541, 585; Ex. 1001, 7:19—
`
`22). Petitioner also explains that “it would have been obvious to have a flow
`
`key additionally include a next-hop MAC layer address” “in order to verify
`
`the flow key against the information relevant to the connection during
`
`lookup.” Pet. 49 (citing Ex. 1003 111] 30—31, A-14; Ex. 1006, .585). We also
`
`credit Petitioner’s expert’s uncontested testimony that “[a] next-hop MAC
`
`layer address indicates the physical location a packet is first routed on its
`
`way to its final destination for the connection and is therefore information
`
`relevant to the connection.” Pet. 49 (citing Ex. 1003, 30—31, A-14).
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`Patent Owner argues that Petitioner fails to “identify any disclosure in
`
`the prior art of a flow key that includes a first hop MAC,” provide “anything
`
`more than an unadomed conclusion,” provide a “stated rationale in the prior
`
`art for adding additional information, let alone a next-hop MAC address
`
`specifically,” or “explain why a flow key that includes a first hop MAC
`
`address would have been obvious.” PO Resp. 27—29. We are not persuaded
`
`by Patent Owner’s argument at least because Patent Owner does not assert
`
`or demonstrate specific flaws in the rationale provided by Petitioner.
`
`For example, as Petitioner and Petitioner’s expert explain, when
`
`“Ethernet is used,” one of skill in the art would have known that “source and
`
`destination MAC (media access control) addresses” “determine[] the next
`’9 “
`
`hop along the route to the destination
`
`such as a network interface card
`
`(NIC).” See, e. g., Ex. 1003 1111 28, 30, 31. Hence, Petitioner’s evidence
`
`demonstrates that one of skill in the art would have known that transmission
`
`of data in a network to a destination, such as to a network interface card, is
`
`accomplished by a known “MAC address” that identifies a destination of the
`
`data. We agree with Petitioner that it would have been obvious to one of
`
`ordinary skill in the art, when attempting to transmit data in a network to a
`
`destination, to include a known parameter (i.e., a MAC address) that
`
`indicates the destination of the transmitted data. Otherwise, the data would
`
`not be transmitted to the desired destination, the desired destination not
`
`having been identified by the (known) MAC address.
`
`Claim 32 recites an “operation code.” Petitioner argues that the
`
`combination of Thia and Tanenbaum discloses this feature. Pet. 52—55. For
`
`example, Petitioner argues that Thia discloses “checking the header portion
`
`of an incoming packet to see if it matches what is expected or predicted,”
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`subsequently making a “fast-path determination,” then “rout[ing] the packet
`
`.
`
`.
`
`. for fast-path/bypass processing.” Pet. 52—53 (citing Ex. 1015, .003,
`
`.006, .007, .013, .014, Table 1, Fig. 2; Ex. 1006, .585). Petitioner also
`
`explains that it would have been obvious to one of ordinary skill in the art to
`
`“us[e] an operation code .
`
`.
`
`. instructing whether to process the packet on
`
`the” bypass interface by, in one example, “checking the header portion of an
`
`incoming packet to see if it matches,” or, in another example, “a flag .
`
`.
`
`.
`
`signifying that a packet can be fast-pathed.” Pet. 52—55 (citing Ex. 1015,
`
`.004, .009, A-22—A-24).
`
`Patent Owner argues that “Petitioners tacitly admit there is no explicit
`
`disclosure in Thia or Tanenbaum of the claimed ‘operation code’,” that
`
`“Thia does not use an ‘operation code’ for the routing of PDUs,” “does not
`
`need to also associate the PDU with an operation code,” and only discloses a
`
`“PDU header” that supposedly does not include “a component .
`
`.
`
`. to
`
`determine whether it is bypassable.” PO Resp. 30, 32, 39—40. To the extent
`
`that Patent Owner argues that Thia fails to refer to matching a header to
`
`indicate whether the packet is a candidate for avoiding processing by the
`
`host computer (i.e., fast-path processing) as an “operation code,” we are not
`
`persuaded by Patent Owner’s argument at least because this is not an
`
`“ipsissimis verbis” test. In re Bond, 910 F.2d 831, 832 (Fed. Cir. 1990).
`Patent Owner does not explain a sufficient difference between matching a
`
`header to determine if a packet is a candidate for avoiding processing by the
`
`host computer (i.e., fast-path processing) of Thia and the claimed “operation
`
`code,” that is also for determining if a packet is a candidate for avoiding
`
`processing by the host computer.
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`To the extent that Patent Owner argues that an “operation code” is
`
`“associated” with a data packet by indicating a status of the packet, as
`
`recited in claim 32, for example, we agree with Petitioner that Thia discloses
`
`this feature for at least the previously discussed reasons. For example, Thia
`
`discloses indicating that a packet is bypassable by “match[ing] the incoming
`
`PDU headers with a template that identifies the predicted bypassable
`
`headers.” Ex. 1015, .003. One of skill in the art would have understood that
`
`the matching of the header of the packet to a template would have been
`
`“associated” with the packet itself. Otherwise, the header of the packet
`
`would not have been matched with the template, the packet itself having no
`
`“association” with the matching.
`
`Also, as Petitioner points out, Thia discloses,
`
`The “no—in-transit PDU” test can often be avoided. At the
`beginning of data transfer on a new connection,
`it
`is
`automatically satisfied.
`It holds as long as no packet fails a
`bypass test, and it is sufficient to maintain a flag to indicate this.
`Once a packet fails, and goes to the SPS, then a full “no—in-transit
`PDU” test must be performed for each packet until the test
`succeeds, after which control can go back to the flag.
`
`Ex. 1015, .003. Thus, Thia’s flag is used to indicate whether any packet
`
`“fails a bypass test.” Once one packet fails the bypass test, requiring
`
`processing by the host protocol stack (SPS), a more complete test (full no-in-
`
`transit PDU) must be performed on each packet until a next packet passes
`
`the full test and the quicker bypass test can be resumed. Therefore, ’l‘hia’s
`
`flag indicates that the long “no-in-transit PDU” test may be avoided for
`
`packets, in favor of the quicker bypass test, until a packet fails the quick
`
`bypass test. The ordinarily skilled artisan would have understood Thia’s
`
`flag is associated with a packet. More precisely, Thia’s flag appears to be a
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`global flag associated with the processing of all received packets rather than
`
`a unique flag associated only with a single corresponding received packet.
`
`However, the claims don’t preclude such a global association with all
`
`received packets.
`
`Patent Owner argues that Thia fails to provide a.“disclosure or
`
`suggestion of using an operation code to call the BYPASS_START
`
`procedure” and that doing so “would have been superfluous and unnecessary
`
`in View of Thia’s architecture.” PO Resp. 38. Even assuming Patent
`
`Owner’s contention to be correct that Thia fails to disclose using an
`
`operation code to call a “BYPASS_START procedure,” we are still not
`
`persuaded by Patent Owner’s implied argument that Thia therefore somehow
`
`also fails to disclose indicating, with an operation code, whether a packet is a
`candidate for avoiding processing by the host computer (i.e., fast-path
`
`processing). As previously discussed, Thia discloses this feature.
`
`Petitioner argues that the combination of Thia and Tanenbaum
`
`discloses storing the header portion of the packet in a header buffer if the
`
`header conforms to the TCP protocol, as recited in claim 32. Pet. 58—60.
`
`For example, Petitioner argues that one of skill in the art would have
`
`understood “that a buffer is a portion of memory” and that “a packet .
`
`.
`
`.
`
`includes header and data portions” and would have been “placed into a
`
`packet memory.” Pet. 58—59 (citing Ex. 1003 A-30; Ex. 1015 Fig. 4).
`
`Patent Owner argues that the combination of 'l‘hia and Tanenbaum
`
`fails to disclose storing a header portion of a packet in a header buffer
`
`because, according to Patent Owner, Thia discloses (in Fig. 4 of Thia)
`
`storing a packet header in a “Protocol Header” portion of memory but that
`
`the portion of memory (i.e., “Protocol Header”) that stores the header
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`portion of a packet “are not the claimed separate ‘header buffer.” PO Resp.
`
`41; Ex. 1015 Fig. 4. We are not persuaded by Patent Owner’s arguments at
`
`least because Patent Owner does not point out sufficient differences between
`
`the portion of memory that stores a header portion of a packet of Thia and
`
`the “header buffer,” as recited in claim 32, for example. In both cases, a
`
`header portion of a packet is stored in a specified portion of memory.
`
`Patent Owner also argues that Thia fails to disclose storing the header
`
`portion of the packet “if said header conforms to the TCP protocol.” PO
`
`Resp. 42. However, we are persuaded by Petitioner and Petitioner’s expert
`
`that one of ordinary skill in the art, having ordinary creativity and not being
`
`an automaton, would have understood that, when performing a process of
`
`“fast—path” processing for TCP, the header would conform to the TCP
`
`protocol. Otherwise, one of skill in the art would have understood that the
`
`process would not operate as desired, the header not conforming to the
`
`protocol in use in the system.
`
`Petitioner argues that the combination of Thia and Tanenbaum
`
`discloses a “flow re-assembler” that re-assembles a data portion of a packet
`
`with a data portion of another packet, as recited in claims 41 and 42, for
`
`example, that Tanenbaum “suggests that the re-assembler is provided on a
`
`network interface,” and that Thia “provides an express disclosure.” Pet.
`
`77—78 (citing Ex. 1006, .498). In particular, Petitioner argues that
`Tanenbaum “discloses that the TCP protocol processes packets and re-
`
`assembles the data portions” and that “packets conforming to the TCP
`
`protocol are processed and the data portions .
`
`.
`
`. are re-assembled in the
`
`correct order together into a buffer.” Pet. 57—5 8, 77.
`
`10
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`Patent Owner argues that “Petitioners admit Tanenbaum does not
`
`disclose a flow re—assembler,” as recited in claim 41, but fails to demonstrate
`
`that Petitioner, in fact, “admitted” that Tanenbaum fails to disclose the
`
`disputed claim limitation or how the re-assembly of data portions of packets
`
`of Tanenbaum (relied upon by Petitioner) differs from the reassembly of
`data portions of packets, as recited in claim 41. In both cases, data portions
`
`of packets are reassembled. PO ReSp. 43.
`
`Petitioner argues that the combination of Thia and Tanenbaum
`
`discloses a processor that maintains a TCP connection that is stored as a
`
`control block on the network interface, as recited in claims 41 and 42. Pet.
`
`78—8 1. For example, Petitioner argues that “[t]his limitation reflects
`
`standard TCP protocol operation .
`
`.
`
`. on a network interface” in which “a
`
`packet is received [and] classified .
`
`.
`
`. [and] processed” including “updating
`
`the TCP connection for the communication flow.” Pet. 78. In addition,
`Petitioner argues that Thia discloses “control blocks to store information” by
`
`“sending information .
`
`.
`
`. to the bypass chip” that are “stored in a process
`
`control block.” Pet. 79 (citing Ex. 1015, .009).
`
`Patent Owner argues that Thia and Tanenbaum fail to disclose a
`
`“processor.” PO Resp. 45, 50. In the absence of sufficient evidence to
`
`demonstrate persuasively that one of skill in the art, given the level of skill
`
`in the art at the time of the invention, would not have understood that a
`
`“processor” is used to process data, we agree with the Petitioner that one of
`
`skill in the art being of ordinary creativity and not being an automaton would
`
`have understood that a “processor” would have been utilized in a system that
`
`“processes” packets (as disclosed by the combination of Thia and
`
`Tanenbaum).
`
`11
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`Patent Owner does not provide additional argument with respect to the
`
`other challenged claims.
`
`Combinabilig
`Petitioner argues that it would have been obvious to one of ordinary
`
`skill in the art to have combined the teachings of Thia and Tanenbaum. Pet.
`
`30—32. In particular, Petitioner argues that Thia discloses that one of
`
`ordinary skill in the art would have known and understood “offloading
`
`network protocol processing,” “would have recognized that 081 and TCP/1P
`
`share many similarities,” and that the “architecture of [the] bypass
`
`implementation” is used for “any standard protocol.” Pet. 31—32; Ex. 1015,
`.003; Ex. 1003 1] 110; Ex. 1006, 056—057.
`V
`
`Petitioner also argues that Tanenbaum discloses a similar system for
`
`“expedited network protocol processing similar to Thia” including “the
`
`concept of Header Prediction” and that “many TCP implementations use it.”
`
`Pet. 33 (citing Ex. 1006, 584—585). Hence, as Petitioner points out, it
`
`would have been obvious to one of ordinary skill in the art to have combined
`
`the known process of “header prediction” in “fast-path processing” with
`
`“protocol processing of layered protocols” using “any standard protocol’
`
`(i.e., Thia) with the known system of “header prediction” in “fast-path
`
`processing” (i.e., Thia or Tanenbaum) using “TCP implementations” (i.e.,
`
`Tanenbaum) as a known “standard protocol” (i.e., Thia and Tanenbaum) to
`
`achieve the predictable and expected result of a system for “fast-path
`
`processing” using “any standard protocol” such as “TCP implementations.”
`
`We are persuaded by Petitioner’s arguments. “The combination of familiar
`
`elements according to known methods is likely to be obvious when it does
`
`12
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`no more than yield predictable results.” KSR Int ’1 Co. v. Teleflex, Inc., 550
`
`US. 398, 416 (2007).
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the cited references because
`
`“Tanenbaum teaches away from performing protocol processing using
`
`anything’other than the host CPU.” PO Resp. 54 (citing Ex. 1006, .588—
`
`.589). However, the cited portion of Tanenbaum discloses that if an effort is
`
`made to “avoid having the network coprocessor be as expensive as the main
`
`CPU, it is often a slower chip,” which results in the “(fast) CPU [being] idle
`
`waiting for the second (slow) CPU to do the critical work.” Ex. 1006, .588—
`
`.589. Hence, Tanenbaum actually discloses that the system may not be
`
`optimal if a less “expensive” CPU is selected and the “slow CPU” “do[es]
`
`the critical work,” which is directed to the cost of the CPU used and not
`
`whether a “host CPU” is used or not. We are not persuaded by Patent
`
`Owner’s argument.
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the cited references because
`
`“Tanenbaum explains the lack of interest in OSI.” PO Resp. 55. However,
`
`as previously discussed, Petitioner relies on Tanenbaum for disclosing
`
`TCP/IP and not OSI. Even assuming Patent Owner’s contention to be
`
`correct that Tanenbaum supposedly discloses a “lack of interest in OSI,”
`
`Patent Owner does not assert or demonstrate persuasively that this alleged
`
`disclosure regarding an supposed “lack of interest in OSI” sufficiently
`
`refutes Petitioner’s prima facie showing of obviousness of the disputed
`
`claims over the combination of Thia and Tanenbaum.
`
`13
`
`
`
`IPR2017—01410
`
`Patent 8,131,880 B2
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the cited references because Thia
`
`discloses that “bypass could be used ‘for any standard protocol’ .
`
`.
`
`. [but
`
`actually means] any standard OSI protocol.” PO Resp. 55. We are not
`
`persuaded by Patent Owner’s argument at least because Patent Owner does
`
`not provide sufficient evidence supporting Patent Owner’s allegation that
`
`one of skill in the art would have understood that Thia intended to disclose
`
`“any OSI protocol” but inadvertently discloses “any standard protocol.” We
`
`agree with Petitioner (Pet. 31—36 (citing Ex. 1003)) that one of ordinary skill
`
`in the art, when confronted with the phrase “any standard protocol,” would
`
`have understood the phrase to mean “any standard protocol” and would have
`
`not instead understood the phrase to mean something else — namely, “any
`
`OSI protocol.” Even assuming Thia discloses “any standard OSI protocol”
`
`(Thia does not disclose this limitation, however), Patent Owner does not
`
`sufficiently demonstrate that one of skill in the art would have understood
`
`“any standard OSI protocol” to also mean “but not the TCP/IP protocol.”
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the cited references because Thia
`
`discloses “modify[ing] existing OSI stack software” but fails to disclose “a
`
`TCP/IP protocol stack.” PO Resp. 57. We note that Thia discloses a system
`
`that is “based on .
`
`.
`
`. a generalization of Jacobson's "Header Prediction"
`
`algorithm .
`
`.
`
`. for TCP/IP.” Ex. 1015, .002. Patent Owner does not explain
`
`how Thia’s system that is based on “TCP/IP” actually does not use a
`
`“TCP/IP protocol stack.” In any event, even assuming Patent Owner’s
`
`contention to be correct that Thia supposedly fails to disclose TCP/IP, Patent
`
`Owner does not assert or demonstrate persuasively that the combination of
`
`14
`
`
`
`1PR2017-01410
`
`Patent 8,131,880 B2
`
`Thia and Tanenbaum also fails to disclose TCP/IP. As previously discussed,
`
`the combination of Thia and Tanenbaum discloses this feature.
`
`Secondagy Considerations
`
`Patent Owner also argues that it would not have been obvious to one
`
`of ordinary skill in the art to have combined the teachings of Thia and
`
`Tanenbaum because there was a “long-felt but unsolved need” “to enhance
`
`the efficiency of network protocol processing and network traffic
`
`management” and that “[t]he nexus between the long-felt need and the
`
`claimed invention” is to “solve[]” “bottlenecks.” PO Resp. 58—59. We are
`
`not persuaded by Patent Owner’s argument for at least the reasons set forth
`
`by Petitioner. Pet. Reply 20—21. We agree with Petitioner that Patent
`
`Owner has not persuasively established any connection between resolution
`
`of those bottlenecks and the patented invention. To be accorded substantial
`
`weight, there must be a nexus between the claimed invention and the
`
`evidence of secondary considerations. In re GPAC Inc, 57 F.3d 1573, 1580
`
`(Fed. Cir. 1995). Nexus is a legally and factually sufficient connection
`
`between the objective evidence and the claimed invention, such that the
`
`objective evidence should be considered in determining nonobviousness.
`
`Demaco Corp. v. F. von LangsdorflLicensing Ltd., 851 F.2d 1387, 1392
`(Fed. Cir. 1988). The burden of showing that there is a nexus lies with the
`
`Patent Owner. See Paulsen, 30 F.3d at 1482. In the absence of an
`
`established nexus with the claimed invention, secondary consideration
`
`factors are not entitled to much, if any, weight and generally have no bearing
`
`on the legal issue of obviousness. See In re Vamco Mach. & Tool, Inc, 752
`
`F.2d 1564, 1577 (Fed. Cir. 1985). Moreover, to the extent that Patent Owner
`
`argues that there was a “long-felt need” to solve “bottlenecks,” Patent
`
`15
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`Owner does not assert or demonstrate persuasively that any of the disputed
`
`claims recites “solving bottlenecks.” To the extent that Patent Owner argues
`
`that some unspecified limitation recited in claim 32 constitutes the required
`
`“nexus” to the alleged “long-felt need,” we note that Thia previously
`
`satisfied this need. The “long-felt need” must not have been satisfied by
`
`another before the patentee. Newell Co. v. Kenney Mfg. Co., 864 F.2d 757,
`
`768 (Fed. Cir. 1988).
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to haVe combined the teachings of the cited
`
`references because “the challenged claims .
`
`.
`
`. enjoyed great commercial
`
`success” by “the offloading .
`
`.
`
`. technology described in the challenged
`
`claims.” PO Resp. 60. We are not persuaded by Patent Owner’s argument
`
`for at least the reasons set forth by Petitioner. Pet. Reply 21—22. Patent
`
`Owner does not provide sufficient information or evidence to establish that
`
`the claimed invention, in fact, experienced “commercial success.” In fact, as
`
`Petitioner argues, evidence of record indicates that the claimed invention
`
`“never went anywhere” and was ultimately “deprecated.” Pet. Reply 22
`
`(citing Exs. 1224, 1227, 1228, 1230). In any event, even assuming that the
`
`claimed invention experienced “commercial success,” as Patent Owner
`
`alleges, the feature Patent Owner alleges to have resulted in the presumed
`
`“commercial success” was previously disclosed by Thia. See discussion
`
`above. Under these circumstances, any alleged commercial success stems
`
`from what was known in the prior art so that there can be no nexus. Tokaz'
`
`Corp. v. Easton Enters, Inc., 632 F.3d 1358, 1369 (Fed. Cir. 2011).
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the teachings of the cited
`
`16
`
`
`
`IPR2017-Ol410
`
`Patent 8,131,880 B2
`
`references because “Alacritech’s patent portfolio covering network
`
`acceleration techniques was the subject of several successful commercial
`
`licenses.” PO Resp. 60. We are not persuaded by Patent Owner’s
`
`arguments for at leaSt the reasons set forth by Petitioner. Pet. Reply 22—23.
`
`For example, as Petitioner explains, Patent Owner does not demonstrate
`sufficiently that the alleged licenses were the result of the claimed invention
`
`and, therefore, fails to establish a nexus between the claimed invention and
`
`the alleged licenses. See, e. g., Pet. Reply 22—23. Rather, as Petitioner points
`
`out, the licenses were the result of reasons not related to the claimed
`
`invention (e.g., as a result of an infringement lawsuit). Pet. Reply 23 (citing
`
`Ex. 2038). In any event, even assuming that there were “successful
`
`commercial licenses,” as Patent Owner contends, and the alleged “successful
`
`commercial licenses” were the result of some unspecified feature recited in
`
`claim 32, for example, as previously discussed, Thia discloses these features.
`
`There can be no nexus if the feature relied upon was previously known in
`
`the prior art. Tokaz' Corp, 632 F.3d at 1369.
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the teachings of the cited
`
`references because the claimed invention was alleged to be the subject of
`
`industry “praise.” PO Resp. 61—62. We are not persuaded by Patent
`
`Owner’s argument for at least the reasons set forth by Petitioner. Pet. Reply
`
`23. For example, Patent Owner argues that various sources stated that
`
`Patent Owner’s network interface card “is able to sustain network
`
`bandwidth,” “achiev[es] lower processor utilization,” and “is an
`
`evolutionary advancement of [Patent Owner’s] .
`
`.
`
`. protocol acceleration”
`
`(PO Resp. 61—62 (citing Ex. 2039 11 4; Ex. 2026 11 189; Ex. 2026 11 190)), but
`
`17
`
`
`
`IPR2017-01410
`
`Patent 8,131,880 B2
`
`Patent Owner does not demonstrate sufficiently that any of these alleged
`
`statements, assuming that any of these statements would have been
`
`considered to be “praise” at all, pertain to the claimed invention and in what
`
`way. Hence, Patent Owner fails to establish sufficient nexus between the
`
`alleged “praise” and the claimed invention.
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the teachings of the cited
`
`references because “prior attempts at ‘TCP offload [have] repeatedly
`
`failed.’” PO Resp. 62 (citing Ex. 2041, 001—013). We are not persuaded by
`
`Patent Owner’s argument for at least the reasons set forth by Petitioner. Pet.
`
`Reply 24. Even if TCP offload is a form of network processing offload, the
`
`Patent Owner provides no evidence linking the failure of others to any
`
`limitations of the challenged claims. Also, as Petitioner points out, Thia
`
`itself discloses a “generalization of the ‘Header Prediction’ algorithm for
`
`TCP/1P” and that “its teachings are compatible with ‘any standard
`
`protocol.”’ Pet. Reply 2—3. Patent Owner states that “TCP offload”
`
`supposedly “repeatedly failed” but does not explain sufficiently how a
`
`system (of Thia) that is based upon an “algorithm for TCP/1P” and -
`
`applicable to “any standard protocol” (which one of skill in the art would
`
`have understood to include TCP because, at least, the Thia system is a
`
`generalization of such a system) would have failed. We note that Thia does
`
`not disclose that its generalized “TCP/1P” offload system fails.
`
`Patent Owner argues that it would not have been obvious to one of
`
`ordinary skill in the art to have combined the teachings of the cited
`
`references because “experts and industry were skeptical of offloading
`
`processing of complex protocols.” PO Resp. 63. We are not persuaded by
`
`18
`
`
`
`IPR2017-01410
`Patent 8,131,880 B2
`
`Patent Owner’s argument for at least the reasons set forth by Petitioner. Pet.
`
`Reply 24. For example, as previously discussed, Thia, for example,
`
`discloses “offloading processing of complex protocols.” There can be no
`
`nexus if the feature relied upon was previously known in the prior art. Tokaz'
`
`Corp, 632 F.3d at 1369. Nor would one of ordinary skill in the art have
`
`been “skeptical” of procedures (e.g., offloading) already disclosed in the
`
`prior art (e. g., Thia).
`
`Real Parties in Interest
`
`Intel Corporation identifies itself as a real party in interest in these
`
`proceedings and represents that “[n]o other parties exercised or could have
`
`exercised control over this Petition; no other parties funded or directed this
`
`Petition.” Pet. 2. Patent Owner argues that “Dell is both Intel’s and
`
`Cavium’s customer and indemnitee,” that “Del