UNITED STATES PATENT AND TRADEMARK OFFICE
`______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`
`INTEL CORP., CAVIUM LLC, and DELL INC.,
`
`Petitioner,
`
`v.
`
`ALACRITECH, INC.,
`
`Patent Owner.
`______________________
`Case IPR2018-002341
`U.S. Patent No. 8,805,948
`______________________
`REPLY TO PATENT OWNER’S RESPONSE TO PETITION
`FOR INTER PARTES REVIEW OF U.S. PATENT NO. 8,805,948
`
`
`Mail Stop “PATENT BOARD”
`Patent Trial and Appeal Board
`U.S. Patent and Trademark Office
`P.O. Box 1450
`Alexandria, VA 22313-1450
`
`
`
` 1
`
` Cavium LLC (formerly Cavium, Inc.), which filed a Petition in Case IPR2018-
`
`00403, and Dell Inc., which filed a Petition in Case IPR2018-01307, have been
`
`joined as petitioners in this proceeding.
`
`
`______________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`______________________
`
`INTEL CORP., CAVIUM LLC, and DELL INC.,
`
`Petitioner,
`
`v.
`
`ALACRITECH, INC.,
`
`Patent Owner.
`______________________
`Case IPR2018-002341
`U.S. Patent No. 8,805,948
`______________________
`REPLY TO PATENT OWNER’S RESPONSE TO PETITION
`FOR INTER PARTES REVIEW OF U.S. PATENT NO. 8,805,948
`
`
`Mail Stop “PATENT BOARD”
`Patent Trial and Appeal Board
`U.S. Patent and Trademark Office
`P.O. Box 1450
`Alexandria, VA 22313-1450
`
`
`
` 1
`
` Cavium LLC (formerly Cavium, Inc.), which filed a Petition in Case IPR2018-
`
`00403, and Dell Inc., which filed a Petition in Case IPR2018-01307, have been
`
`joined as petitioners in this proceeding.
`
`
`
`I.
`II.
`
`TABLE OF CONTENTS
`
`THE COMBINATION DISCLOSES A NETWORK INTERFACE
`
`INTRODUCTION .......................................................................................... 1
`PO MISCHARACTERIZES THE PRIOR ART ......................................... 1
`A.
`Thia ....................................................................................................... 1
`B.
`Tanenbaum96 ....................................................................................... 3
`C.
`Stevens2 ................................................................................................ 3
`III. A POSA WOULD BE MOTIVATED TO COMBINE THIA AND
`TANENBAUM96 .......................................................................................... 4
`IV. A POSA WOULD HAVE BEEN MOTIVATED TO COMBINE
`THIA, TANENBAUM96, AND STEVENS2 ............................................. 7
`V.
`CHECKING WHETHER PACKETS ARE IP FRAGMENTED .............. 8
`VI. THE COMBINATION TEACHES CHECKING WHETHER THE
`“PACKETS” HAVE CERTAIN EXCEPTION CONDITIONS ............. 10
`VII. THE COMBINATION DISCLOSES THE PROTOCOL STACK
`PROCESSING OF EXCEPTION CONDITIONS LIMITATIONS ........ 12
`VIII. THE COMBINATION DISCLOSES THE BYPASSING HOST
`LIMITATIONS ............................................................................................ 12
`IX. THE EVIDENCE OF OBVIOUSNESS FAR OUTWEIGHS PATENT
`OWNER’S ALLEGED “OBJECTIVE EVIDENCE” .................................. 14
`A.
`Of The 948 Patent And The “Objective Evidence” ......................... 14
`B.
`There Is No Evidence of Commercial Success ................................... 15
`C.
`Failing, or Skepticism ......................................................................... 16
`
`PROTOCOL STACK PROCESSING AND STORING DATA
`FROM PACKETS WITHOUT EXCEPTION CONDITIONS
`
`PO Has Not Shown A Nexus Between The Challenged Claims
`
`There Is No Evidence of Long-Felt Need, Praise, Trying and
`
`i
`
`
`
`X.
`
`Intel Filed This Petition within One-Year of Being Accused of
`
`THE PETITION IS NOT TIME BARRED UNDER § 315(B) .................... 16
`A.
`Infringement in the Underlying Litigations ........................................ 17
`B.
`The Defendants Are Not RPIs for this Proceeding ............................. 18
`1.
`Intel is the Sole RPI Under the Applicable Standard ............... 18
`2.
`Insufficient to Make Defendants RPIs ...................................... 19
`3.
`PO Overextends Recent Federal Circuit Case Law .................. 21
`Intel and Defendants are Not in Privity ............................................... 23
`C.
`D.
`Doctrines of RPI and Privity ............................................................... 25
`XI. CONCLUSION .............................................................................................. 26
`
`The Relationship Between Intel and Defendants is
`
`The Facts Do Not Justify the Application of the Equitable
`
`
`
`
`
`
`
`ii
`
`
`
`Cases
`
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Applications in Internet Time, LLC v. RPX Corp.,
`897 F.3d 1336 (Fed. Cir. 2018) ............................................................................ 27
`ARM Ltd v. AMD, Inc.,
`IPR2018-01148, Paper 16 (PTAB Dec. 12, 2018) ....................................... 28, 29
`
`Bosch Auto. Serv. Sols., LLC v. Matal,
`878 F.3d 1027 (Fed. Cir. 2017) ............................................................................ 25
`Geo. M. Martin Co. v. Alliance Mach. Sys. Int'l LLC,
`618 F.3d 1294 (Fed. Cir. 2010) .............................................................................. 2
`Google v. Seven Networks,
`IPR2018-01047, Ex. 1056 (PTAB Dec. 3, 2018) ......................................... 29, 31
`MRC Innovations, Inc. v. Hunter Mfg., LLP,
`747 F.3d 1326 (Fed. Cir. 2014) ............................................................................ 24
`Polaris Indus., Inc. v. Arctic Cat, Inc.,
`882 F.3d 1056 (Fed. Cir. 2018) ............................................................................ 11
`St. Jude Medical v. Volcano Corp.,
`Case No. IPR2013-00258, Paper 29 (PTAB Oct. 16, 2013) ............................... 26
`Taylor v. Sturgell,
`553 U.S. 880 (2008) ................................................................................ 32, 33, 36
`
`Tokai Corp. v. Easton Enters., Inc.,
`632 F.3d 1358 (Fed. Cir. 2011) ............................................................................ 24
`Unified Patents, Inc. v. Realtime Adaptive Streaming, LLC,
`IPR2018-00883, 2018 WL 6504233 (PTAB Oct. 11, 2018) ............................... 31
`WesternGeco LLC v. ION Geophysical Corp.,
`889 F.3d 1308 (Fed. Cir. 2018) ..................................................................... 33, 34
`Wi-Fi One, LLC v. Broadcom Corp.,
`887 F.3d 1329 (Fed. Cir. 2018) .................................................................... passim
`
`iii
`
`
`
`Worlds Inc. v. Bungie, Inc.,
`903 F.3d 1237 (Fed. Cir. 2018) ............................................................................ 30
`
`Wyers v. Master Lock Co.,
`616 F.3d 1231 (Fed. Cir. 2010) ............................................................................ 23
`
`Statutes
`
`35 U.S.C. § 315(b) ............................................................................................ 16, 19
`
`
`
`
`
`iv
`
`
`
`Exhibit #
`1224
`
`1227
`
`1228
`1230
`
`1232
`
`1399
`
`1411
`
`1412
`
`1413
`
`1414
`
`1415
`
`1417
`
`1418
`
`TABLE OF EXHIBITS
`
`
`Description
`Deposition of Kevin C. Almeroth, Ph.D., Volume 1 (May 03,
`2018)
`New ASIC drives Alacritech into storage by R. Merritt, EE
`Times (January 11, 2011)
`Internet page from Alacritech.com downloaded on May 6, 2018
`Why Are We Deprecating Network Performance Features? By
`B.Wilson downloaded on May 2, 2018
`Alacritech’s First Amended and Supplemental Patent Initial
`Disclosures for Intel from Alacritech v. CenturyLink, et al.,
`16cv693, Eastern District of Texas (February 24, 2017)
`Declaration of Robert Horst, Ph. D. in Support of Petitioner’s
`Reply to Patent Owner’s Response to Petition for Inter Partes
`Review of U.S. Patent 8,805,948
`Order Granting Intel Corporation’s Motion to Intervene from
`Alacritech Inc. v. CenturyLink Communications LLC, et al.,
`Case No. 2:16-cv-00693-JRG-RSP (November 21, 2016)
`Alacritech’s Answer and Counterclaims to Intel Corporation’s
`Complaint in Intervention from Alacritech Inc. v. CenturyLink
`Communications LLC, et al., Case No. 2:16-cv-00693-JRG-
`RSP (December 13, 2016)
`Stipulation and Joint Motion to Stay Litigation Pending IPR
`Proceedings from Alacritech Inc. v. CenturyLink
`Communications LLC, et al., Case No. 2:16-cv-00693-JRG-
`RSP (December 4, 2017)
`Declaration of Garland T. Stephens
`(Petitioner’s Restricted – Attorneys’ Eyes Only)
`Intel Corporation’s Motion to Intervene from Alacritech Inc. v.
`CenturyLink Communications LLC, et al., Case No. 2:16-cv-
`00693-JRG
`Alacritech’s Response to Intel Corporation’s Motion to
`Intervene from Alacritech Inc. v. CenturyLink Communications
`LLC, et al., Case No. 2:16-cv-00695-JRG-RSP (November 17,
`2016)
`Alacritech’s Response to Intel Corporation’s Motion to
`Intervene from Alacritech Inc. v. CenturyLink Communications
`
`v
`
`
`
`Exhibit #
`
`1419
`
`1420
`
`1422
`
`1423
`
`1424
`
`1440
`
`Description
`LLC, et al., Case No. 2:16-cv-00695-JRG-RSP (March 3,
`2017)
`Declaration of Karineh Khachatourian in Support of Cavium
`LLC’s Opposition to Alacritech, Inc.’s Motion for Additional
`Discovery from Case No. IPR2018-00401
`Article: Fragmentation Considered Harmful by C. Kent and J.
`Mogul (October 1987)
`Agreement between Intel Corporation and Dell Computer
` (88819DOC000001- 88819DOC000003)
`(Petitioner’s Restricted – Attorneys’ Eyes Only)
`Amendment to Agreement between Intel Corporation and Dell
`Computer (88819DOC000004-88819DOC000006)
`(Petitioner’s Restricted – Attorneys’ Eyes Only)
`
`Intel Terms and Conditions of Sale
`(88819DOC000007-88819DOC000012)
`(Petitioner’s Restricted – Attorneys’ Eyes Only)
`
`Deposition of Robert Horst in Case No. IPR2018-00234
`
`vi
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`
`I.
`
`INTRODUCTION
`PO’s Response (Paper 18, “POR”) relies on incorrect views of the prior art
`
`and improperly assumes a POSA requires verbatim teachings to build the claimed
`
`subject matter.
`
`In reality, the claims recite well-known TCP/IP operations—including header
`
`prediction and fast-path processing—where the simple TCP fast-path processing is
`
`offloaded to external hardware. As set forth in the Petition, Tanenbaum96 and
`
`Stevens2 teach TCP/IP and fast-path/bypass processing, and Thia explicitly teaches
`
`the benefits of offloading fast-path/bypass processing to hardware and therefore
`
`invalidate the challenged claims.
`
`II. PO MISCHARACTERIZES THE PRIOR ART
`A. Thia
`PO argues that “Thia discloses … an inoperative device, and is a non-enabling
`
`reference,” and lacks details “required to enable the implementation.” POR at 13.2
`
`The Board previously rejected PO’s position. See 1409 FWD at 8.3 Operability is
`
`
`
` 2
`
` All emphasis is added unless otherwise indicated.
`
`3 PO recycles the same arguments regarding Thia and Tanenbaum96 previously
`
`rejected in Final Written Decisions for the related 880 Patent. See, e.g., IPR2017-
`
`01409, Paper 79 at11-14 (“1409 FWD”).
`
`1
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`irrelevant to the obviousness analysis. See, e.g., Geo. M. Martin Co. v. Alliance
`
`Mach. Sys. Int'l LLC, 618 F.3d 1294, 1302-03 (Fed. Cir. 2010);
`
`PO overlooks Thia’s teachings and ignores a POSA’s understanding in view
`
`of Tanenbaum96. For example, PO alleges that Thia’s bypass is shown only in
`
`Figure 1 (Ex. 2026 ¶86), ignoring Thia’s many bypass disclosures. See, e.g., Ex.
`
`1015 at .003, .004); Horst Reply Decl. (“Ex. 1399”) ¶¶26-28.
`
`PO also overlooks that Thia’s bypass “is a generalization of Jacobson’s
`
`‘Header Prediction’ algorithm … for TCP/IP” and that the receive bypass test
`
`identifies bypassable headers. Id. at .002, .003. Thia’s bypassable packets are those
`
`in the data-transfer phase for an OPEN connection. Ex. 1015 at .003, .004.
`
`Tanenbaum96 discloses that TCP/IP processing is “straightforward” in the data-
`
`transfer (ESTABLISHED) state. Ex. 1006 at .583. PO’s expert offers no reason why
`
`a POSA could not implement a bypass test from these disclosures. Moreover,
`
`Tanenbaum96 teaches predictable headers for bypass/fast-path processing. Id. at
`
`.585; see also Ex. 1013 at .962 (explaining that “a small set of tests” determines if a
`
`packet can be fast-path processed). Additionally, Tanenbaum96 and Stevens2, like
`
`Thia, cite Van Jacobson’s “header prediction”. See Ex. 1006 at .585 (“Many TCP
`
`implementations use it”); Ex. 1013 at .960 (“Header prediction was put into the
`
`4.3BSD Reno release); Ex. 1015 at .002 (“generalization of Jacobson’s ‘Header
`
`Prediction’ algorithm … for TCP/IP”). A POSA would have understood how to
`
`2
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`implement Thia’s receive bypass test for TCP/IP, as well as Thia’s and
`
`Tanenbaum96’s teachings. See Ex. 1003 ¶¶99-122; Ex. 1399 ¶28.
`
`PO makes two additional mischaracterizations. First, PO argues that the host
`
`reassembles data from PDUs into larger data blocks, but does not point to any
`
`applicable disclosures in Thia. POR at 17. As Petitioner explained (see, e.g., Paper
`
`2 (“Petition”) at 80-83), Thia discloses re-assembly of data by the ROPE chip.
`
`Second, PO asserts Thia is limited to OSI. However, Thia is applicable to “any
`
`standard protocol.” See Section III.
`
`B. Tanenbaum96
`PO argues that Tanenbaum96 teaches away from hardware bypass. But PO
`
`misconstrues Tanenbaum96, contending “Tanenbaum teaches that integration of
`
`transport entity functions in the NIC is complicated and non-trivial.” POR at 20
`
`(citing Ex. 2026 ¶¶ 102-103). However, here Tanenbaum96 is discussing the entire
`
`“transport entity”, not the simple transport entity “functions” (e.g., TCP/IP fast-path)
`
`relevant here. See Section III; see also 1409 FWD at 9-10 (rejecting PO’s argument).
`
`C.
`Stevens2
`PO argues that “Stevens does not address offloading protocol processing to a
`
`network interface chip,” (POR at 22) but Stevens2 is not cited for this proposition.
`
`Thia and Tanenbaum96 address offloading protocol processing. Petition at 40-46,
`
`53, 56-58.
`
`3
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`III. A POSA WOULD BE MOTIVATED TO COMBINE THIA AND
`TANENBAUM96
`A POSA would have been motivated to combine Thia and Tanenbaum96. See
`
`Petition at 56-58. PO’s recycled arguments to the contrary were rejected. See, e.g.,
`
`1409 FWD at 11-14.
`
`First, PO alleges Tanenbaum96 teaches away from the combination. POR at
`
`44-45 (citing Ex. 1006 at .588- .589). But Tanenbaum96’s opinion concerns
`
`offloading a complex protocol stack, not the streamlined bypass/fast-path taught by
`
`Thia, Tanenbaum96, or Stevens2. See Ex. 1006 at .588-.589 (“unless the protocol is
`
`exceedingly simple ….”); Ex. 1399 ¶¶48-50, 56. Tanenbaum96 also teaches that
`
`TPDU processing is “straightforward” in the ESTABLISHED state. See Ex. 1006
`
`at .583.
`
`Consistent with Tannenbaum96, Thia’s “Reduced Operation Protocol
`
`Engine” chip only handles packets that do not require complex processing. See Ex.
`
`1015 at .001, .003. Thia’s bypass comprises “relatively simple operations needed for
`
`data-transfer across multiple layers” performed in hardware. Id. at .002. A POSA
`
`would have understood that Tanenbaum96 does not negate Thia’s teachings. Thia
`
`recognizes the concerns with hardware offloads identified by Tanenbaum96 and
`
`provides a solution. Compare Ex. 1015 at .002 (“Probably because of the complexity
`
`of existing protocols … implementation … has been disappointing …. Bypass solves
`
`the problems identified above, which may limit the use of offboard processing ….”)
`
`4
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`with Ex. 1006 at .588 (“[U]nless the protocol is exceedingly simple, hardware just
`
`means a plug-in board with a second CPU ….”). A POSA would be motivated to
`
`combine Tanenbaum96’s TCP/IP teachings with Thia’s solutions. Ex. 1015 at .013;
`
`Ex. 1399 ¶50.
`
`Second, PO asserts that TCP/IP’s increasing popularity over OSI cuts against
`
`a motivation to combine. POR at 45-46. PO’s assertions rest on the faulty
`
`assumption that a POSA would not look to Thia’s offload teachings because it
`
`discloses an OSI implementation. Id. It is the opposite—a POSA would have
`
`combined Thia’s bypass architecture with the popular TCP/IP protocol. Petition at
`
`26, 56.; Ex. 1399 ¶51.
`
`Third, PO contends Thia’s teaching that its bypass architecture is “for any
`
`standard protocol” is limited to the OSI protocol. POR at 46-47. But, Thia’s
`
`disclosure applies generally to multi-layered protocols (like TCP/IP). See Ex. 1015
`
`at .001.005 (“the bypass system can be extended to incorporate multiple-layer
`
`stacks.”); Ex. 1399 ¶¶52-53. Rather than limiting Thia to the OSI protocol, Thia
`
`describes a generic “standard protocol stack (SPS)”. Ex. 1015 at .003. Moreover,
`
`Thia’s bypass stack is a generalization of the “‘Header Prediction’ algorithm … for
`
`TCP/IP”. Id. at .002.
`
`PO does not explain why Thia’s architecture is inapplicable to TCP/IP and
`
`merely points to one of Thia’s cited papers as evidence that Thia is restricted to OSI.
`
`5
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`POR at 47; Ex. 2026 ¶141. But this paper confirms that Thia’s bypass concept is not
`
`so limited. See Ex. 1038 at .001 (“A protocol bypass is a fast processing path” that
`
`can work with “standardized layered protocol system like OSI …”), .002, .015.
`
`PO also suggests that Thia’s disclosure of presentation and session layers
`
`limits it to OSI. POR at 46-47. But Petitioner’s expert explained a POSA would have
`
`understood the relationship between these layers and TCP/IP. Ex. 1003 ¶¶22-28.
`
`Moreover, the 948 Patent acknowledges that these layers extend to the TCP/IP
`
`protocol. Ex. 1001 at 2:10-20 (“… TCP/IP … integrates the functions of session
`
`layers and presentation layers ….”); see also Ex. 1399 ¶¶53-54.
`
`PO also argues “[n]either FDDI nor ATM utilize the TCP/IP protocol.” POR
`
`at 47. However, using TCP/IP with FDDI was well-known. See Ex. 1252 at .022
`
`(depicting TCP and IP layers on top of various layers, including FDDI).
`
`Finally, PO argues that Thia’s use of an existing OSI stack discourages a
`
`POSA from using TCP/IP. POR at 48. Thia’s use of “an existing implementation of
`
`the OSI stack” to “provid[e] an easy migration path for current systems” does not
`
`imply that modifying other protocol stacks would not be easy or within a POSA’s
`
`skills. Ex. 1015 at .014. See also Ex. 1399 ¶55. Here, Thia’s general preference for
`
`an existing stack does not teach away because it does not criticize, discredit, or
`
`otherwise discourage using another protocol like TCP/IP. Polaris Indus., Inc. v.
`
`6
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`Arctic Cat, Inc., 882 F.3d 1056, 1069 (Fed. Cir. 2018). Instead, Thia teaches its
`
`architecture is for multi-layer protocols, such as TCP/IP.
`
`IV. A POSA WOULD HAVE BEEN MOTIVATED TO COMBINE
`THIA, TANENBAUM96, AND STEVENS2
`PO raises three flawed arguments why there is no motivation to combine Thia,
`
`Tanenbaum96, and Stevens2.
`
`First, PO argues that Stevens2’s header prediction is limited to host processes.
`
`POR at 49; Ex. 2026 ¶145. But PO fails to identify any supporting disclosure in
`
`Stevens2. Further, a POSA would have been motivated to apply Stevens2’s TCP/IP
`
`header prediction to Thia’s bypass test for its fast-path offload, which is based on
`
`TCP/IP header prediction. See Petition at 56-60; Ex. 1399 ¶56.
`
`Second, PO references “Tanenbaum’s teaching away from implementing
`
`protocol processing in hardware” (POR at 49), but as explained above,
`
`Tanenbaum96 does not teach away from offloading the simple protocol processing.
`
`See Section III, supra; Ex. 1399 ¶56.
`
`Third, PO argues, without expert support, that the combination is the result of
`
`improper hindsight because Thia “illustrates that even professors … were uncertain
`
`of the feasibility of offloading ….” POR at 49-50. However, Thia concluded that
`
`“[t]he design is practical” (Ex. 1015 at .001) and that “it is feasible to implement the
`
`bypass stack” (id. at .013). Further, Dr. Horst explains that Thia’s gate-level design
`
`7
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`could be used to fabricate a chip to verify its teachings. See Ex. 1399 ¶56; Ex. 1015
`
`at .008.
`
`V. THE COMBINATION DISCLOSES A NETWORK INTERFACE
`CHECKING WHETHER PACKETS ARE IP FRAGMENTED
`PO disputes that the combination teaches checking received packets for
`
`certain exception conditions, “including whether the packets are IP fragmented.”
`
`POR at 24-30; Ex. 2026 ¶¶110-120. This argument fails.
`
`The combination teaches a hardware fast-path offload for TCP/IP. Petition at
`
`56-61, 75. Thia’s receive bypass test checks “PDU headers” to determine whether
`
`packets are bypassable (Ex. 1015 at .003) and Tanenbaum96 teaches checking for
`
`fragmentation in TCP/IP (Ex. 1006 at .585). Petition at 75. PO does not address these
`
`disclosures and fails to rebut that Thia combined with the TCP/IP teachings of
`
`Tanenbaum96 and Stevens2 teach a receive bypass test that checks incoming packet
`
`headers (not TPDU headers) for the absence of IP fragmentation (i.e., Thia checks
`
`for packets with conditions that are outside the data transfer state; Tanenbaum96 and
`
`Stevens2 teach TCP/IP for the fast-path; and Tanenbaum96 checks for a full (not
`
`fragmented) TPDU to identify bypassable packets). See Petition at 75.
`
`PO also alleges that Tanenbaum96’s checking whether “the TPDU is a full
`
`one” (Ex. 1006 at .585) “is distinct from checking whether the packets are IP
`
`fragmented.” POR at 24; Ex. 2026 ¶111. However, PO does not dispute that a full
`
`8
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`TPDU is one that is not IP fragmented. Ex. 1003 ¶114. Checking whether a “TPDU
`
`is a full one” confirms whether the packet is IP fragmented and satisfies the claim.
`
`Additionally, it is undisputed that the combination teaches a receive bypass
`
`test that screens fragmented packets. See Ex. 1015 at .014. Whether fragmentation
`
`can be detected without checking the network layer is irrelevant. There are many
`
`ways a POSA could check whether incoming packets are fragmented. Dr. Horst
`
`suggests that one can validate the checksum in the TCP layer. See Ex. 1440 at 20:21-
`
`23. PO does not disagree with Dr. Horst (or argue it would not be obvious), but
`
`argues it is not the best way. POR at 26-28; Ex. 2026 ¶¶114-115. This is insufficient
`
`to overcome the prior art disclosures.
`
`Further, PO admits “a POSA would recognize that a network interface would
`
`check whether a packet is IP fragmented by checking IP headers ….” POR at 26;
`
`Ex. 2026 ¶114. Dr. Almeroth agrees, “[a] network interface can quickly check IP
`
`headers for IP fragmentation and determine whether packets should bypass the host
`
`protocol processing ….” Ex. 2026 ¶114. Given Thia’s teaching of checking packet
`
`headers to determine bypassability, Tanenbaum96’s disclosure of confirming “the
`
`TPDU is a full one”, and PO’s admission that a POSA would know how to check
`
`for fragmentation, the combination teaches a network interface checking whether
`
`received packets have certain conditions, “including whether the packets are IP
`
`fragmented.”
`
`9
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`VI. THE COMBINATION TEACHES CHECKING WHETHER THE
`“PACKETS” HAVE CERTAIN EXCEPTION CONDITIONS
`PO contends the combination does not disclose checking “whether the
`
`
`
`packets have certain exception conditions,” because Tanenbaum96 discloses
`
`checking the “TPDU” (Ex. 1006 at .585) and Stevens2 examines a “segment” (Ex.
`
`1013 at .962-.963). See POR at 31-38. At the outset, a POSA would not have
`
`understood the claimed “packet” is limited to an IP packet. As Dr. Horst explains,
`
`“packet” can refer to protocol data units at different protocol layers, as the 948
`
`Patent confirms. See Ex. 1399, ¶36; Ex. 1001 at 10:57-61 (referring to a TPDU, or
`
`segment, as a “TCP packet”).
`
`
`
`Even if “packet” means an “IP packet,” all of PO’s unsupported arguments
`
`overlook that a TPDU/segment is part of, or encapsulated within, an IP packet.
`
`Ex. 1006 at .503 (red showing a TPDU within a “packet”); Ex. 1003 ¶¶29-30.
`
`
`
`10
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`
`Ex. 1006 at .542. Due to encapsulation, a characteristic of a TPDU is also that of
`
`the IP packet. This is consistent with the claims, which recite “checking whether
`
`the packets have a FIN flag [(a TPDU header field)] set.” Ex. 1001, claim 1.
`
`Checking whether the TPDU header in an IP packet has a FIN flag set is checking
`
`whether the IP packet has a FIN flag set. See Petition at 75; Ex. 1399 ¶¶34-35.
`
`
`
`
`
`11
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`
`Ex. 1006 at .544.
`
`
`
`PO suggests that the claims require checking only the network layer header
`
`of a packet. POR at 31. But the Institution Decision correctly determined this is not
`
`the case. Paper 7 at 27. The claims encompass checking multiple layers or a single
`
`layer.4
`
`
`
`Even if checking the network layer header is required—it is not—header
`
`prediction does this. Tanenbaum96 discloses that header prediction begins by
`
`parsing the IP header of the received packet for the source and destination IP
`
`addresses, which are used to look up the connection record. Ex. 1006 at .584-.585.
`
`VII. THE COMBINATION DISCLOSES THE PROTOCOL STACK
`PROCESSING OF EXCEPTION CONDITIONS LIMITATIONS
`PO’s arguments that these limitations are not disclosed are based entirely on
`
`
`
`arguments in pages 24-38 of its Response. See POR at 38-39. These arguments fail
`
`for the same reasons explained above in Section VI.
`
`VIII. THE COMBINATION DISCLOSES THE BYPASSING HOST
`PROTOCOL STACK PROCESSING AND STORING DATA
`
`
`
` 4
`
` Even if PO were correct that the claims require checking a packet at its network
`
`layer, the Petition discusses Thia’s receive bypass test (Petition at 75), which checks
`
`“incoming PDU headers.” Ex. 1015 at .003.
`
`12
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`FROM PACKETS WITHOUT EXCEPTION CONDITIONS
`LIMITATIONS
`PO has failed to rebut the Petition’s showing that the prior art discloses these
`
`
`
`limitations. Petition at 78-83, 92; Ex. 1399 ¶¶37-45.
`
`PO alleges that these limitations are not disclosed based in part on its
`
`arguments in pages 24-38 of its Response. See POR at 39. These arguments fail for
`
`the same reasons explained above in Section VI. Further, the Board found that the
`
`combination of Thia and Tanenbaum96 teaches storing data on the host without
`
`TCP headers. 1409 FWD at 10-11.
`
`
`
`PO’s remaining arguments do not rebut the combination. First, PO argues
`
`the combination does not disclose storing payload data on the host without TCP
`
`headers in between because Thia allegedly transfers a whole PDU to be processed
`
`by the host. POR at 40 (citing Ex. 1015 at .009); Ex. 2026 ¶124. But PO citation in
`
`Thia discusses transmitting packets from the host, not receiving packets. Ex. 1015,
`
`Thia at .009; Ex. 1399 ¶¶38-39.
`
`
`
`Second, PO asserts that Thia’s disclosure of header decoding in the bypass
`
`and that “a data portion of a packet may be copied between the host and the
`
`network interface adapter …” is insufficient because it doesn’t disclose that the
`
`header is not also copied. POR at 40-41. However, conventional practice is to strip
`
`off the headers after protocol processing. Ex. 1003 at ¶38. Thia’s teaching is
`
`13
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`consistent with this practice—transferring the data portion to the host from the
`
`ROPE chip after the packet has been processed. See Ex. 1015 at .003, .007.
`
`
`
`Third, PO is incorrect that Tanenbaum96’s teaching that TCP “reconstructs
`
`the original byte streams” (Ex. 1006 at .540) fails to teach this limitation and offers
`
`no explanation for its argument. POR at 41.
`
`IX. THE EVIDENCE OF OBVIOUSNESS FAR OUTWEIGHS PATENT
`OWNER’S ALLEGED “OBJECTIVE EVIDENCE”
`A. PO Has Not Shown A Nexus Between The Challenged Claims Of
`The 948 Patent And The “Objective Evidence”
`
`PO makes no effort to tie any of the alleged “objective evidence”—including
`
`
`
`PO’s products—to the features of the challenged claims. See Wyers v. Master Lock
`
`Co., 616 F.3d 1231, 1246 (Fed. Cir. 2010). This is unsurprising given that PO stated
`
`it was not relying on its products practicing the claimed inventions in the 948 Patent
`
`in the co-pending litigations. Ex. 1412; Ex. 1232.005-.006. PO also does not show
`
`that any of the “objective evidence” stems from the invention, and not the prior art.
`
`Tokai Corp. v. Easton Enters., Inc., 632 F.3d 1358, 1369 (Fed. Cir. 2011). PO’s
`
`unsupported statements fail to establish nexus.5 MRC Innovations, Inc. v. Hunter
`
`
`
` 5
`
` The Board has already found PO failed to provide evidence of nexus for related
`
`patents based on the same arguments and “objective evidence.” IPR2017-01405
`
`FWD at 10-14; compare POR at 50-57 with IPR2017-01405, Paper 31 at 54-61.
`
`14
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`
`Mfg., LLP, 747 F.3d 1326, 1336 (Fed. Cir. 2014).
`
`A. There Is No Evidence of Commercial Success
`PO has provided no evidence that its products, Petitioner’s products, or other
`
`products practice the claims of the 948 Patent. POR at 52. Furthermore, PO’s
`
`products were not commercially successful. Ex. 1227; Ex. 1228.001. Its products
`
`offered a “TCP Offload Engine” (“TOE”). Ex. 1227.001. As its own CEO, Larry
`
`Boucher, admitted, TOE “never went anywhere,” and PO “only had limited fringe
`
`of people trying to get performance” with TOE. Ex. 1227.001. Microsoft disabled
`
`Chimney in Windows and then subsequently deprecated it. Ex. 1230.003. Despite
`
`this, PO’s expert still claims that PO’s products enjoyed “great commercial success.”
`
`Ex. 1224 at 208:18-209:5.
`
`PO also argues that the inventions were subject to “several successful
`
`commercial licenses” but provides no evidence of nexus. POR at 52-53. These
`
`licenses resulted from a lawsuit that PO filed against Microsoft in 2004 asserting
`
`that Microsoft’s software and Broadcom’s hardware were infringing different
`
`patents on TOE technology. Ex. 2038 at 272-287. A handful of other makers of
`
`hardware NICs also took licenses to TOE technology so they could utilize the
`
`Microsoft Chimney software that supported TOE. Id. at 1-31, 62-135, 171-204, 215-
`
`271. PO has not provided any evidence that any of these licenses were driven by any
`
`15
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`challenged claim of the 948 Patent. See, e.g., Bosch Auto. Serv. Sols., LLC v. Matal,
`
`878 F.3d 1027, 1038 (Fed. Cir. 2017).
`
`B.
`
`There Is No Evidence of Long-Felt Need, Praise, Trying and
`Failing, or Skepticism
`PO provides no evidence that the alleged long-felt need relates to the
`
`challenged claims. POR at 50-52. Its conclusory statement from its expert is
`
`insufficient to establish nexus. For alleged praise, PO relies solely on two articles
`
`related to its products (POR at 53-54), but has provided no evidence that its products
`
`practiced the challenged claims. For alleged trying and failing, PO provides no
`
`evidence of nexus between the single article it cites and the challenged claims. POR
`
`at 54-55.
`
`PO’s alleged evidence of skepticism similarly lacks any nexus. Moreover,
`
`PO’s misconstrues the disclosures. Tanenbaum96 never suggests a “plug-in board
`
`with a second CPU and its own program” will not work with a fast-enough processor,
`
`and Dr. Horst’s article in fact confirms that the “conventional wisdom” was that
`
`special purpose NICs were used for TCP/IP acceleration. Ex. 2300 at 1.
`
`X. THE PETITION IS NOT TIME BARRED UNDER § 315(B)
`Dell, Wistron, or CenturyLink (“Defendants”) are not RPIs in this IPR and
`
`PO does not allege that they are privies of Intel. They are not. Whether a non-party
`
`is a RPI or privy for purposes of an IPR proceeding is an equitable doctrine that is a
`
`“highly fact-dependent question” that takes into account how courts generally have
`
`16
`
`
`
`Case IPR2018-00234
`U.S. Patent No. 8,805,948
`used the terms to “describe relationships and considerations sufficient to justify
`
`applying conventional principles of estoppel and preclusion.” Office Patent Trial
`
`Practice Guide (“TPG”), 77 Fed. Reg. 48756, 48759 (Aug. 14, 2012).
`
`A.
`
`Intel Filed This Petition within One-Year of Being Accused of
`Infringement in the Underlying Litigations
`PO sued Defendants on June 30, 2016 for patent infringement, alleging that
`
`Intel components in the Defendants’ accused products meet some asserted claim
`
`limitations. The Court granted Intel’s Motion to Intervene on Nov. 21, 2016. Ex.
`
`1411. On December 13, 2016, PO served counterclaims accusing Intel’s products of
`
`infringing patents-in-suit, including the 948 Patent, establishing Intel’s one-year bar
`
`date of December 13, 2018. Ex. 1412; see St. Jude Medical v. Volcano Corp., Case
`
`No. IPR2013-00258, Paper 29 at 2-3 (PTAB Oct. 16, 2013) (counter-claims
`
`establish one year bar).
`
`Intel filed IPR2017-01395 (“Original 948 Petition”) within the Defendants’
`
`one-year window. The present Petition, filed within two weeks of the Board’s
`
`decision declining to institute the Original 948 Petition and within Intel’s one-year
`
`window, is nearly identical apart from additional evidence establishing that SMB
`
`was publicly available. Paper 7 at 14. The grounds in thi
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
