`IPR2015-00172
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`APPLE INC., HTC CORPORATION, HTC AMERICA, INC., SAMSUNG
`ELECTRONICS CO. LTD, SAMSUNG ELECTRONICS AMERICA, INC.,
`SAMSUNG TELECOMMUNICATIONS AMERICA, LLC AND
`AMAZON.COM, INC.,
`Petitioners
`
`v.
`
`MEMORY INTEGRITY, LLC
`Patent Owner
`
`U.S. Patent No. 7,296,121
`
`
`
`Inter Partes Review Case No. 2015-00172
`
`
`
`MEMORY INTEGRITY, LLC’S PATENT OWNER
`PRELIMINARY RESPONSE PURSUANT TO 37 CFR § 42.107(a)
`
`
`
`
`
`
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`TABLE OF CONTENTS
`
`INTRODUCTION ........................................................................................... 1
`
`TECHNOLOGY BACKGROUND ................................................................. 1
`
`
`
`
`I.
`
`II.
`
`III. SUMMARY OF PETITIONERS’ PROPOSED GROUND FOR
`REVIEW .......................................................................................................... 2
`
`IV. THE PENDING PETITIONS FOR INTER PARTES REVIEW OF THE
`’121 PATENT PRESENT REDUNDANT GROUNDS ................................. 3
`
`V. MEMORY INTEGRITY, LLC’S CLAIM CONSTRUCTIONS ................. 10
`
`A.
`
`B.
`
`“probe filtering unit” (claims 1, 16, 25) ................................................ 11
`
`“accumulate responses to each probe” and “accumulating probe
`responses” (claims 15 and 25) ............................................................... 13
`
`VI. THERE IS NO REASONABLE LIKELIHOOD OF PETITIONERS
`PREVAILING AS TO A CHALLENGED CLAIM OF THE ’121
`PATENT ........................................................................................................ 14
`
`A. Petitioners Failed to Demonstrate That Stanford-DASH Anticipates
`Claims 1-3, 8, 11, 12, 16, 19, 20 and 22 ............................................... 14
`
`1.
`
`2.
`
`Petitioners Fail to Demonstrate that Stanford-DASH
`Anticipates Any Independent Claim Because Stanford-DASH
`Does Not Disclose a “probe filtering unit” “operable to filter
`probes within a single cluster of processing nodes” ..................... 14
`
`Petitioners Fail to Demonstrate That the “directory board” of a
`“home cluster” in Stanford-DASH “transmit[s] the” received
`“probes” “only to selected ones of the processing nodes” “with
`reference to probe filtering information” ...................................... 21
`
`
`
`i
`
`
`
`3.
`
`4.
`
`5.
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Petitioners Also Fail to Demonstrate that Stanford-DASH
`Anticipates Claims 2 or 3 Because Stanford-DASH’s
`“directory board” is not “an additional node interconnected
`with the plurality of processing nodes via the first point-to-
`point architecture” ......................................................................... 25
`
`Petitioners Also Fail to Demonstrate that Stanford-DASH
`Anticipates Claim 8 Because the “directory board” for a “local
`cluster” in Stanford-DASH Always “transmits” a “request” to
`the “local cluster” Before It “transmits” the “request” to the
`“home cluster” ............................................................................... 27
`
`Petitioners Fail to Demonstrate that Stanford-DASH
`Anticipates Claims 11 or 12 Because Stanford-DASH
`Contains No Disclosure Regarding the Programming of its
`Processors ...................................................................................... 29
`
`B. Petitioners Failed To Demonstrate That Claims 4-6 Are Obvious
`Over Stanford DASH In View of Keller ............................................... 31
`
`C. Petitioners Failed To Demonstrate That Claim 7 Is Obvious Over
`Stanford-DASH In View of HyperTransport ........................................ 34
`
`D. Petitioners Failed To Demonstrate That Claim 9 Is Obvious Over
`Stanford DASH In View of Duato ........................................................ 36
`
`E. Petitioners Failed To Demonstrate That Claims 17-24 Are Obvious
`Over Stanford-DASH In View of Smith ............................................... 39
`
`VII. CONCLUSION .............................................................................................. 40
`
`
`
`ii
`
`
`
`
`
`
`
`
`
`Exhibit No.
`Memory Integrity-2001
`
`Memory Integrity-2002
`
`Memory Integrity-2003
`
`Memory Integrity-2004
`
`Memory Integrity-2005
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`EXHIBIT LIST
`
`Description
`Plaintiff Memory Integrity, LLC’s Initial Identification
`of Asserted Claims And Accused Products, served on
`Petitioners in Memory Integrity LLC v. Amazon.com
`Inc., et al., Nos. 1:13-cv-01795, -01796, -01802,
`-01808 (D. Del. served Oct. 13, 2014)
`Excerpts from D. E. Culler, J. P. Singh, and A. Gupta
`PARALLEL COMPUTER ARCHITECTURE, pp. 279-280
`(1999)
`Sorin et al. , “Specifying and Verifying a Broadcast and
`a Multicast Snooping Cache Coherence Protocol,”
`IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED
`SYSTEMS, Vol. 13, No. 6, pp. 1-23(June 2002)
`Excerpts from Merriam-Webster’s Collegiate
`Dictionary (10th ed. 1999)
`
`Excerpts from David A. Patterson, et al., COMPUTER
`ORGANIZATION AND DESIGN (3d ed. 2005)
`
`
`
`iii
`
`
`
`
`
`
`
`TABLE OF AUTHORITIES
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
` Page(s)
`
`Cases
`Canon Inc. v. Intellectual Ventures I LLC,
`Case No. IPR2014-00535 (PTAB Sep. 24, 2014) ................................................ 5
`
`Canon Inc. v. Intellectual Ventures I LLC,
`Case No. IPR2014-00536 (PTAB Nov. 5, 2014) ................................................. 5
`
`Cisco Systems, Inc. v. C-Cation Technologies, LLC,
`Case No. IPR2014-00454 ................................................................................... 28
`
`Crown Operations Int'l, LTD v. Solutia Inc.,
`289 F.3d 1367 (Fed. Cir. 2002) .......................................................................... 30
`
`In re Cuozzo Speed Technologies, LLC,
`No. 2014-1301 (Fed. Cir. Feb. 4, 2015) ......................................................... 5, 11
`
`Gracenote, Inc. v. Iceberg Industries LLC,
`Case No. IPR2013-00551 (PTAB Feb. 28, 2014) .............................................. 12
`
`Illumina, Inc. v. Trustees of Columbia Univ.,
`IPR2012-00006 (PTAB May 10, 2013)............................................................ 4, 8
`
`Leveen v. Edwards,
`57 U.S.P.Q.2d 1406 (B.P.A.I. 2000) .................................................................. 28
`
`Ex parte Levy,
`17 USPQ2d 1461 (Bd. Pat. App. & Inter. 1990) ................................................ 30
`
`Microsoft Corporation v. Surfcast, Inc.,
`Case No. IPR2013-00292 (PTAB Nov. 19, 2013) ............................................... 5
`
`In re Oelrich,
`666 F.2d 578 (C.C.P.A. 1981) ............................................................................ 30
`
`Oracle Corporation v. Clouding IP, LLC,
`IPR2013-0088 (PTAB June 13, 2013) .............................................................. 5, 8
`
`
`
`iv
`
`
`
`Philips v. AWH Corp.,
`415 F.3d 1303(Fed. Cir. 2005) ........................................................................... 11
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Statutes
`
`35 U.S.C. § 312(a)(3)(A) & (a)(5) ........................................................................... 37
`
`Rules
`
`Fed. R. Evid. 1002 ................................................................................................... 37
`
`Other Authorities
`
`32 C.F.R. § 42.107(a) ............................................................................................... 10
`
`37 C.F.R. § 42.1(b) .................................................................................................... 6
`
`37 C.F.R. § 42.22(a)(2) ............................................................................................ 12
`
`37 C.F.R. § 42.62 ..................................................................................................... 37
`
`37 C.F.R. § 42.104(b)(3) .................................................................................... 11, 12
`
`v
`
`
`
`
`
`
`
`I. INTRODUCTION
`The Board should deny the present request for inter partes review of U.S.
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Patent No. 7,296,121 (“the ’121 patent”) because the Petition for inter partes
`
`review is based on an incorrect construction of several claim terms, the Petition
`
`fails to offer a construction of other claim terms, the Petition relies on inadequately
`
`explained inherency theories, the Petition fails to demonstrate that a single
`
`embodiment of Stanford-DASH renders any of the claims anticipated, and the
`
`Petition’s asserted obviousness grounds are incomplete and do not adequately
`
`justify the asserted combinations.
`
`For these reasons as expressed more fully below, the Petitioners have failed
`
`to demonstrate that there is a reasonable likelihood that they will prevail with
`
`respect to at least one of the claims challenged in the petition. Accordingly, the
`
`Board should deny the Petition.
`
`II. TECHNOLOGY BACKGROUND
`The technology of the ’121 patent generally relates to maintaining the
`
`coherency, or consistency, between copies of information stored in caches of a
`
`multiple processor computer system. Ex. 1001 at 1:22-34. Processors often use
`
`small cache memories that the processor is able to read and write to much faster
`
`than main memory. Id. at 1:26-44. Because each processor has a cache memory,
`
`multiple copies of the same data can reside in multiple cache memories. Id. at
`
`
`
`1
`
`
`
`1:35-45. A problem arises when a processor attempts to change the data in the
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`cache memory of the first processor while at the same time another processor also
`
`attempts to change the value of the same data located in another cache memory
`
`because different values for the same data may result. Id. Cache coherency
`
`generally relates to techniques that maintain the consistency of the data stored in
`
`the processors’ cache memories. Id.
`
`In order to maintain consistency across the same data stored in more than
`
`one cache memory, messages can be sent between the cache memories when
`
`information in the cache changes. However, such messages can result in
`
`significant traffic. The ’121 patent is directed at maintaining cache coherency
`
`while reducing the number of messages that need to be sent. Id. at 2:46-52. A
`
`probe filtering unit is connected to the various processing nodes and is configured
`
`to receive probes from the nodes. Id. at 2:52-56. The probe filtering unit uses
`
`information relating to the state of the cache memories in order to determine which
`
`nodes should receive these messages. Id. at 2:52-3:5. The ‘121 Patent thereby
`
`discloses a system that maintains coherency between the various cache memories
`
`while also reducing the number of messages that need to be transmitted. Id.
`
`III. SUMMARY OF PETITIONERS’ PROPOSED
`GROUND FOR REVIEW
`
`For the Board’s convenience below is a summary of the grounds for review
`
`
`
`2
`
`
`
`proposed by Petitioners:
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`1.
`
`Claims 1-3, 8, 11, 12, 16, 19, 20, and 22: Anticipation under Daniel
`
`Lenoski et al., “The Directory-Based Cache Coherence Protocol for
`
`the DASH Multiprocessor,” ISCA '90 Proceedings of the 17th annual
`
`international symposium on Computer Architecture, pp. 148-159
`
`(May 1990) (“Stanford-DASH”);
`
`2.
`
`Claim 4-6: Obviousness over Stanford-DASH in view of U.S. Patent
`
`Number 6,490,661 to Keller et al (“Keller”)
`
`3.
`
`Claim 7: Obviousness over Stanford-DASH in view of
`
`“HyperTransport™ Technology I/O Link - A High-Bandwidth I/O
`
`Architecture” (Jul. 20, 2001) (“HyperTransport”)
`
`4.
`
`Claim 9: Obviousness over Stanford-DASH in view of Jose Duato et
`
`al., INTERCONNECTION NETWORKS -AN ENGINEERING
`
`APPROACH (1997) (“Duato”) and
`
`5.
`
`Claims 17-24: Obviousness over Stanford-DASH in view of Michael
`
`John Sebastian Smith, APPLICATION-SPECIFIC INTEGRATED
`
`CIRCUITS (1997) (“Smith”).
`
`IV. THE PENDING PETITIONS FOR INTER PARTES REVIEW
`OF THE ’121 PATENT PRESENT REDUNDANT GROUNDS
`
`The Board has made clear that in order to ensure “the just, speedy, and
`
`
`
`3
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`inexpensive resolution of every proceeding,” it will not institute inter partes review
`
`proceedings on cumulative or redundant grounds. Illumina, Inc. v. Trustees of
`
`Columbia Univ., IPR2012-00006, Paper 41, at 11-12 (PTAB May 10, 2013). Thus,
`
`the Board has instructed parties that it will not “authorize inter partes review on
`
`certain unpatentability challenges . . . [where] the challenges appeared to rely on
`
`the same prior art facts as other challenges for which inter partes review had been
`
`authorized.” Id. “In other words, considering multiple rejections for the same
`
`unpatentability issue would unnecessarily consume the time and resources of all
`
`parties involved.” Id. Thus, to avoid dismissal of a proposed ground of
`
`unpatentability, a petitioner must “provide a meaningful distinction between the
`
`different, redundant rejections.” Id. Where multiple references have been cited for
`
`the same facts, it is not enough for a petitioner to argue that the cited references are
`
`not identical, or to “speculate[] that in certain publications an element may be more
`
`clearly set forth in one publication rather than another.” Id. Rather, a petitioner
`
`must adequately explain the difference between the references and “how this
`
`difference would impact the unpatentability challenge.” Id. This includes
`
`“articulat[ing] a meaningful distinction in terms of relative strengths and
`
`weaknesses with respect to application of the prior art disclosures to one or more
`
`claim limitations,” as well as “why [one reference] is more preferred for satisfying
`
`some elements, while [another reference] is more preferred for satisfying some
`
`
`
`4
`
`
`
`other elements.” Oracle Corporation v. Clouding IP, LLC, IPR2013-0088, Paper
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`13, at 5 (PTAB June 13, 2013).
`
`Moreover, this Board’s rules against redundant and cumulative grounds
`
`cannot be avoided by filing multiple petitions against the same patent, as
`
`Petitioners have done. Where a petitioner files “multiple challenges to” the claims
`
`of the same patent “across separate petitions,” and “does not address the
`
`duplicative nature of its arguments across Petitions,” the petitions shall be
`
`“considered together” by the Board. Canon Inc. v. Intellectual Ventures I LLC,
`
`IPR2014-00535, Paper 9, at 19 (PTAB Sep. 24, 2014). “Petitioner’s separate fee
`
`payments, [] did not assure them that three separate trials would be instituted.”
`
`Canon Inc. v. Intellectual Ventures I LLC, IPR2014-00536, Paper 11, at 4 (PTAB
`
`Nov. 5, 2014). Thus, when presented with multiple petitions with redundant
`
`challenges to the same claims of the same patent, the Board may elect to only
`
`consider one petition. See id. Alternatively, the Board may elect to eliminate
`
`redundancies across the petitions and consolidate the remainder of the petitions “to
`
`administer the proceedings more efficiently.” Microsoft Corporation v. Surfcast,
`
`Inc., IPR2013-00292, Paper 15, at 2 (PTAB Nov. 19, 2013).
`
`Notwithstanding this Board’s clear directive against submitting cumulative
`
`and redundant grounds, the present petition is one of four petitions for inter partes
`
`review simultaneously filed by the same Petitioners, all challenging the ’121
`
`
`
`5
`
`
`
`Patent: IPR2015-00159, IPR2015-00161, IPR2015-00163, and IPR2015-00172.
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Pet. at 2. Two of these petitions, the ’159 and ’172 Petitions, utilize the maximum
`
`sixty pages permitted for a petition for review, while the other two petitions nearly
`
`reach that maximum with fifty-six pages each. Together, Petitioners’ four petitions
`
`present sixteen grounds for review, involving ten distinct asserted prior art
`
`references in two-hundred and thirty-two pages of briefing, as well as over one-
`
`thousand, seven-hundred pages of exhibits, including a one-hundred and twenty-
`
`three page expert declaration, Ex. 1014. Additionally, also pending before the
`
`Board, is IPR2015-00158, another challenge to the ’121 Patent, filed by parties that
`
`are co-defendants to the Petitioners in pending litigations in District Court. The
`
`’158 Petition shares one primary prior art reference with the ’163 Petition, but the
`
`’158 Petition presents seven grounds for review, as well an additional four distinct
`
`alleged prior art references and an additional fifty-nine pages of briefing and claim
`
`charts and an additional sixteen pages of expert declaration. Plainly, the pending
`
`petitions jeopardize this Board’s goal to “secure the just, speedy, and inexpensive
`
`resolution of every proceeding.” 37 C.F.R. § 42.1(b).
`
`Additionally, the pending petitions challenging the ’121 Patent present
`
`grossly redundant and cumulative grounds for review, as demonstrated by this
`
`chart showing the number of grounds asserted against each claim of the ’121
`
`Patent for each of the pending petitions.
`
`
`
`6
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Claim #
`\ Pet. # 1
`’158 Pet. 2 2 2 0 0 0 0 2 0 0 2 2 0 2 4 2 2 2 2 2 2 2 2 2 4
`’159 Pet. 1 1 1 0 0 0 0 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
`’161 Pet. 1 1 1 0 0 0 0 1 1 0 1 0 0 1 1 1 1 1 2 2 1 2 1 1 1
`’163 Pet. 1 1 1 1 1 1 0 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1
`’172 Pet. 1 1 1 1 1 1 1 1 1 0 1 1 0 0 0 1 1 1 2 2 1 2 1 1 0
`Total 6 6 6 2 2 2 1 6 3 1 6 5 1 4 7 6 6 6 8 8 6 8 6 6 7
`
`25
`24
`23
`22
`21
`20
`19
`18
`17
`16
`15
`14
`13
`12
`11
`10
`9
`8
`7
`6
`5
`4
`3
`2
`
`
`
`In total, the pending petitions challenging the ’121 Patent ask this Board to
`
`make one-hundred and twenty-five determinations of whether a particular ground
`
`necessitates cancellation of a particular claim, an average of five grounds per
`
`claim. The grounds for review are redundant and cumulative both within
`
`individual petitions, and across the petitions.
`
`Petitioners argue that the four petitions filed by them are not redundant
`
`because they “presented only those grounds necessary to sufficiently demonstrate
`
`that each claim of the ’121 is not patentable, having demonstrated how various
`
`teachings address the claims divergently.” IPR2015-00159 Pet. at 59. This is
`
`plainly false. Indeed, none of the pending petitions even attempts to describe the
`
`substance of any of the other pending petitions to compare the different grounds
`
`and prior art asserted between them. Moreover, for the claims with redundant
`
`grounds within a single petition, e.g., claims 19, 20, and 22 in the ’161 and ’172
`
`
`
`7
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`Petitions, Petitioners present no explanation for that redundancy. Thus, Petitioners
`
`fall far short of their burden of “provid[ing] a meaningful distinction between the
`
`different, redundant rejections.” Illumina, IPR2012-00006, Paper 41, at 11-12.
`
`Petitioners also argue that “[t]he petition including rejections based on
`
`Chaiken provides the most direct disclosure of any of the petitions of the features
`
`recited in claim 14” However, this is inadequate. Petitioners do not “articulate a
`
`meaningful distinction in terms of relative strengths and weaknesses with respect
`
`to application of the prior art disclosures to one or more claim limitations,” nor do
`
`they explain “why [one reference] is more preferred for satisfying some elements,
`
`while [another reference] is more preferred for satisfying some other elements.”
`
`Oracle, IPR2013-0088, paper 13, at 5 (June 13, 2013).
`
`Petitioners also argue that “Stanford DASH . . . is the only petition that
`
`includes a rejection of claims 4-6 based on prior art that cannot be antedated
`
`through priority or swearing behind,” that “Pong is the only petition that includes a
`
`rejection of claim 13,” and that “Stanford DASH is the only petition that includes a
`
`rejection of claim 7.” IPR2015-00159 Pet. at 59. However, this reflects that
`
`Petitioners have attempted to introduce additional disputed claims in an attempt to
`
`create the false perspective of meaningful non-redundancy between the petitions.
`
`Indeed, each of the pending petitions arises out of patent infringement litigations
`
`brought by Patent Owner in District Court. In those cases, prior to the filing of any
`
`
`
`8
`
`
`
`of the present petitions for inter partes review, Patent Owner served an “Initial
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Identification of Asserted Claims and Accused Products” on each of the
`
`defendants, including each of the Petitioners, which identified claims 1-3, 8, 11-12,
`
`and 14-25 of the ’121 Patent as being asserted. Ex. 2001. It is telling that none of
`
`the claims that Petitioners identify as being unique to a particular petition were
`
`listed in the Initial Identification of Asserted Claims served in the litigations.
`
`Thus, Petitioners demonstrably are using additional claims as a tactic to attempt to
`
`convince this Board to shoe-horn multiple redundant grounds on the other claims
`
`into the IPR proceedings. The Board should not support such tactics.
`
`Indeed, even if any of Petitioners’ arguments as to non-redundancy were a
`
`legitimate excuse for Petitioners’ grossly redundant and excessive filings, they
`
`would only counsel, at most, permitting institution as to those specific identified
`
`claims on those specific allegedly non-redundant grounds. They are no basis for
`
`instituting all petitions on all grounds, as Petitioners request.
`
`Finally, as set forth in Patent Owner’s preliminary response to each petition,
`
`Patent Owner believes that none of the grounds of unpatentability presented by
`
`Petitioners should be instituted on any claim. However, to the extent that the
`
`Board determines that some of the grounds presented by Petitioners should be
`
`instituted, and is seeking to identify a reasonable basis for choosing among the
`
`various petitions, Patent Owner submits that that the goal of “the just, speedy, and
`
`
`
`9
`
`
`
`inexpensive resolution of every proceeding” will best be served by instituting on
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`the grounds submitted in the ’163 Petition, as that petition shares a primary prior
`
`art reference and similar grounds with those presented in the ’158 Petition, filed by
`
`Petitioners’ co-defendants.
`
`V. MEMORY INTEGRITY, LLC’S CLAIM CONSTRUCTIONS
`Because this preliminary response “is limited to setting forth the reasons
`
`why no inter partes review should be instituted,” 32 C.F.R. § 42.107(a), Memory
`
`Integrity does not at this time propose a construction for each term. However, the
`
`following terms are either manifestly incorrectly construed by Petitioners, or are
`
`wholly ignored by Petitioners. Memory Integrity reserves the right to assert any
`
`construction of any term in any Patent Owner’s response, or in any subsequent
`
`filing in this proceeding, or in any other proceeding.1
`
`
`1 Patent Owner acknowledges that the PTAB has determined that the
`
`broadest reasonable interpretation standard is the appropriate standard for
`
`construing claims of an unexpired patent in an IPR proceeding, and that a panel of
`
`the Federal Circuit has recently affirmed that holding. See In re Cuozzo Speed
`
`Technologies, LLC, No. 2014-1301 (Fed. Cir. Feb. 4, 2015). Nonetheless, Patent
`
`Owner contends that the claims should be construed in accordance the same
`
`standard used by the district courts as articulated by the Federal Circuit in Philips
`
`
`
`10
`
`
`
`A.
`
`“probe filtering unit” (claims 1, 16, 25)
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Despite the fact that “probe filtering unit” is recited in the body of every
`
`independent claim of the ’121 Patent, despite the fact that Petitioners admit that the
`
`“probe filtering unit” is the “focus” of the claims (Pet. at 18), and despite the fact
`
`that Petitioners rely on the “probe filtering unit” to argue that the ’121 Patent is not
`
`entitled to the priority of its earlier application (Pet. at 4 n.1), Petitioners make no
`
`effort to construe the term. Rather, the Petition merely contains fleeting, cursory
`
`mentions of the “probe filtering unit” limitation which fail to substantively address
`
`the construction and application of the meaning of the term in the claims. See Pet.
`
`at 4 n.1, 18, 23, 25-26. This is, in and of itself, sufficient to deny the petition in its
`
`entirety. See 37 C.F.R. § 42.104(b)(3) (“the petition must set forth . . . [h]ow the
`
`challenged claim is to be construed.”); Gracenote, Inc. v. Iceberg Industries LLC,
`
`IPR2013-00551, Paper 6, at 38 (PTAB Feb. 28, 2014) (denying petition for failure
`
`to comply with 37 C.F.R. § 42.104(b)(3)); see also 37 C.F.R. § 42.22(a)(2).
`
`If Petitioners’ failure to substantively address this limitation is not
`
`
`v. AWH Corp., 415 F.3d 1303(Fed. Cir. 2005), and Patent Owner explicitly
`
`preserves this issue in the event that the Federal Circuit takes this issue en banc or
`
`there is some other change in the governing law. Patent Owner maintains that its
`
`proposed constructions are correct under either standard.
`
`
`
`11
`
`
`
`dispositive, Memory Integrity submits that the proper construction of a probe
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`filtering unit requires, at least, “an apparatus operable to filter probes within a
`
`single cluster of processors.” For example, the ’121 Patent describes the probe
`
`filtering unit, stating:
`
`The filtering of probes within a cluster, i.e., local probe filtering,
`may be implemented in systems . . . having a single cluster of
`processors. . . . . In [that case], these functionalities may be
`implemented in a device which will be referred to herein as a
`probe filtering unit (PFU). It should be understood that the use of
`the term “probe filtering unit” or “PFU” in the following discussion is
`not intended to be limiting or exclusive. Rather, any device or object
`operable to perform the described functionalities . . . is within the
`scope of the invention.
`
`Ex. 1001 at 26:36-57; see also Ex. 1001 at 26:58-27:4 (“FIG. 18 is a diagrammatic
`
`representation of a multiple processor system 1800 in which embodiments of the
`
`invention relating to the filtering of probes within a single cluster of processors
`
`may be practiced. System 1800 [includes] probe filtering unit 1830.”); Ex 1001 at
`
`29:31-53 (“FIG. 21 is a diagrammatic representation of a transaction flow in which
`
`local probe filtering is facilitated . . . The memory controller . . . generates a probe
`
`to the probe filtering unit PFU. . . . The PFU, in turn, probes nodes N0 and N2
`
`after it applies its directory lookup and probe filtering algorithm”).
`
`Thus, the specification of the ’121 Patent supports construing “probe
`
`
`
`12
`
`
`
`filtering unit” as requiring “an apparatus operable to filter probes within a single
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`cluster of processors.” This is significant because, as discussed below, Stanford-
`
`DASH does not filter probes sent within a cluster.
`
`B.
`
`“accumulate responses to each probe” and “accumulating probe
`responses” (claims 15 and 25)
`
`For purposes of this proceeding, the Board should construe these terms as
`
`“gather two or more responses to a probe.” This construction is supported by the
`
`claim language itself which uses the plural “responses” to indicate that it involves
`
`more than just a single response. Memory Integrity’s construction is also
`
`supported by the specification of the ’121 Patent which describes an example
`
`where the probe filtering unit sends a probe to two nodes and then gathers the
`
`responses from both nodes. In particular, the specification explains that:
`
`The PFU, in turn, probes nodes N0 and N2 after it applies its directory
`lookup and probe filtering algorithm. . . . The PFU then accumulates
`the responses from nodes N0 and N2 and sends two responses (one of
`which may be a read response from N0 or N2) back to the requesting
`CPU.
`Ex. 1001 at 29:42-50. Additionally, the dictionary definition of
`
`“accumulate” is “to gather or pile up esp. little by little.” Ex. 2004 at 8. This is
`
`entirely consistent with Memory Integrity’s proposed construction. The Petitioners
`
`did not propose any construction for this term. Accordingly, the Board should
`
`adopt Memory Integrity’s proposed construction.
`13
`
`
`
`
`
`VI. THERE IS NO REASONABLE LIKELIHOOD OF PETITIONERS
`PREVAILING AS TO A CHALLENGED CLAIM OF THE ’121 PATENT
`A.
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`Petitioners Failed to Demonstrate That Stanford-DASH Anticipates
`Claims 1-3, 8, 11, 12, 16, 19, 20 and 22
`1.
`
`Petitioners Fail to Demonstrate that Stanford-DASH Anticipates
`Any Independent Claim Because Stanford-DASH Does Not
`Disclose a “probe filtering unit” “operable to filter probes within
`a single cluster of processing nodes”
`
`A “probe filtering unit” is a limitation of every independent claim of the
`
`’121 Patent, including challenged claims 1 and 16, and their dependents. As
`
`discussed above, the specification of the ’121 Patent makes clear that the
`
`appropriate construction of “probe filtering unit” is “an apparatus operable to filter
`
`probes within a single cluster of processing nodes.”
`
`However, Stanford-DASH simply does not have a “probe filtering unit” as
`
`properly construed because in Stanford-DASH, a “bus-based snoopy scheme is
`
`used to keep caches coherent within a cluster.” Ex. 1005 at 1 (emphasis added).
`
`That is, Stanford-DASH does not perform filtering within a single cluster of
`
`processing nodes. Petitioners contend that the “directory board” of a “home
`
`cluster” in Stanford-DASH constitutes a “probe filtering unit.” Pet. at 22, 24.2
`
`However, Stanford-DASH expressly teaches that:
`
`
`2 A “home cluster” in Stanford-DASH is the cluster which contains the main
`
`memory and directory for a given physical memory address.” Ex. 1005 at 4.
`
`
`
`14
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`The directory does not maintain information concerning whether the
`home cluster itself is caching a memory block because all
`transactions that change the state of a memory block are issued
`on the bus of the home cluster, and the snoopy bus protocol keeps the
`home cluster coherent.
`
`Ex. 1005 at 5. Stanford-DASH continues, explaining that this is an intentional
`
`design choice to “issue all transactions on the home cluster’s bus”:
`
`While we could have chosen not to issue all transactions on the home
`cluster’s bus this would had an insignificant performance
`improvement since most requests to the home also require an access
`to main memory to retrieve the actual data.
`
`Ex. 1005 at 5. Thus, a home cluster of Stanford-DASH will always probe all
`
`processors in that cluster by transmitting probes over a bus without any filtering.
`
`Therefore, Petitioners’ arguments that the directory board of a home cluster in
`
`Stanford-DASH is a “probe filtering unit” fail because the home cluster issues “all
`
`transactions” on the bus without filtering. Thus, it cannot be an apparatus operable
`
`to filter probes within a single cluster of processors.
`
`Stanford-DASH further emphasizes that all requests to the home cluster are
`
`broadcast to all processors in the home cluster in discussing requests. Petitioners
`
`identify two type of requests in Stanford-DASH as being the allegedly filtered
`
`probes: “read requests” and “read-exclusive requests.” Pet. at 22; Ex. 1014 ¶ C-5.
`
`In both cases, Stanford-DASH expressly states that those requests are broadcast on
`15
`
`
`
`
`
`the home cluster’s bus. Ex. 1005 at 5 (“When the read request reaches the home
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`cluster, it is issued on that cluster’s bus.”); id. at 6 (“At the home cluster, the read-
`
`exclusive request is echoed on the bus.”).
`
`Indeed, that Stanford-DASH’s apparatus is incapable of being a probe
`
`filtering unit is further reinforced by Stanford-DASH’s Figure 3—the “Directory
`
`board block diagram”—which shows how remote requests are processed (i.e. such
`
`as would be received by a home cluster).
`
`
`
`16
`
`
`
`Patent No. 7,296,121
`IPR2015-00172
`
`
`
`
`Annotation of Figure 3 of Stanford-DASH, Ex. 1005 at 4.
`Thus, all remote requests in Stanford-DASH go through the request network, into
`
`the “Pseudo-CPU,” which “[f]orwards remote CPU request[s] to [the] local
`
`MPBUS.” Ex. 1005 at 4, Fig. 3. As a bus, all of those requests sent on the
`
`MPBUS are transmitted to all