`Trials@uspto.gov
`571-272-7822 Entered: February 7, 2023
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`___________
`
`NETFLIX, INC.,
`
`Petitioner,
`
`v.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`CA, INC.,
`Patent Owner.
`___________
`
`IPR2021-01319
`Patent 7,103,794 B2
`__________
`
`Record of Oral Hearing
`Held: November 10, 2022
`_____________
`
`
`
`Before KRISTEN L. DROESCH, NATHAN A. ENGELS, and
`JULIET MITCHELL DIRBA, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`HARPER BATTS, ESQUIRE
`Sheppard, Mullin, Richter & Hampton, LLC
`1540 El Camino Real
`Menlo Park, CA 94025
`
`
`
`
`ON BEHALF OF PATENT OWNER:
`
`
`DANIEL YOUNG, ESQUIRE
`Adsero IP, LLC d/b/a/ Swanson &
`Bratschun, LLC
`8210 South Park Terrace
`Littleton, CA 80120
`
`The above-entitled matter came on for hearing, by video, on
`Thursday, November 10, 2022, commencing at 10:00 a.m., EDT, at the U.S.
`Patent and Trademark Office, before Chris Hofer, Notary Public.
`
`
`
`
`
`
`
`2
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`
` P R O C E E D I N G S
` - - - - -
`JUDGE ENGELS: Good morning. Welcome to the Patent Trial and
`Appeal Board. This is a virtual hearing for IPR 2021-01319 between
`Petitioner Netflix Inc. and Patent Owner CA, Inc. The challenged patent is
`U.S. patent No. 7,103,794 B2. I'm Judge Nathan Engels. With me on the
`panel today are Judges Kristen Droesch and Juliet Dirba. Let's begin with
`introductions of counsel. If you could make your appearances beginning
`with Petitioner, please.
`MR. BATTS: Yes. Good morning, Your Honor. This is Harper
`Batts from the law firm Sheppard Mullin on behalf of Petitioner Netflix and
`with me here today is Jonathan DeFosse, also from Sheppard Mullin.
`JUDGE ENGELS: Thank you, and for Patent Owner.
`MR. YOUNG: Good morning, Your Honor. Dan Young with
`Adsero, IP on behalf of CA, Inc., and with me is Chad King from King IAN,
`LLC.
`
`JUDGE ENGELS: Thank you very much. Each side will have 60
`minutes to present its arguments today. If you choose, you can reserve some
`of your time for rebuttal and surrebuttal. Petitioner will go first, followed by
`Patent Owner and then any rebuttal time that may be reserved. For the
`clarity of the record and to help everyone involved, if you could please make
`sure you identify your demonstrative exhibits by slide number and give us a
`moment to find the appropriate page as you switch pages. Please do not
`interrupt each other during arguments. If you'd like to make objections or
`have issues to raise, please do so in your own time for response for rebuttal
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
`3
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`or surrebuttal.
`We, as I mentioned, we will begin with Petitioner's counsel and before
`we begin, please let me know do you intend to reserve some time for
`rebuttal?
`MR. BATTS: Yes, Your Honor. I'll reserve 20 minutes for rebuttal.
`JUDGE ENGELS: Mr. Young, you don't have to decide now but do
`you intend to reserve time for rebuttal?
`MR. YOUNG: Yes, Your Honor. I’ll reserve 15 minutes.
`JUDGE ENGELS: Okay. Any questions before again?
`MR. BATTS: No, Your Honor.
`MR. YOUNG: No, Your Honor.
`JUDGE ENGELS: With that, Mr. Batts, if you'd like to, I will begin
`the timer when you're ready.
`MR. BATTS: Thank you, Your Honor. As you mentioned, this is for
`the ‘794 patent IPR. So, I'd like to start on slide 2 of Petitioner's slides.
`This IPR has two instituted grounds. One ground which is a Medin plus
`Seltzer combination ground for claims 1, 3 through 9 and 11 through 16 and
`a second ground, ground 2 for claim 17 which is a combination of Medin
`and Markatos, and on slides 3, 4 and 5 we have laid out the three
`independent claims that are being challenged in this IPR and I'd like to
`actually turn to slide 5, claim 17 to start. Given that the only remaining
`asserted claim in the District Court that remains to be challenged here that
`hasn't been invalidated already is claim 17, I want to prioritize my
`presentation with claim 17 and then I'm going to go back and discuss the
`other claims.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`4
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`JUDGE ENGELS: Well, and when you say has not been invalidated
`already, what do you mean by that?
`MR. BATTS: The remaining claims that are being asserted in the
`District Court have been found indefinite by the District Court. So, claim 17
`is the only claim that has basically not been found invalid in some fashion
`already, so that's the claim that I want to prioritize for my client and then I'll
`certainly go back and explain why the remaining claims are also invalid
`under the prior art.
`JUDGE ENGELS: Just to be clear, is that the District Court's
`determination regarding maximizing and minimizing?
`MR. BATTS: Correct, Your Honor.
`JUDGE ENGELS: Thank you. And I'm sorry. Which slide number
`did you?
`MR. BATTS: I’m on slide 5 for claim 17 for ground 2. And so,
`claim 17 is a short claim. It discusses and covers a request for network
`objects and then maintaining the objects in the network in a cache memory
`where the maintaining is independent of a file system. So, for ground 2 the
`Patent Owner has not disputed the receiving step. Patent Owner only
`disputes two things regarding the maintaining, one that the combination
`doesn't maintain network objects in a cache memory and we're going to get
`into the discussion of cache memory, and second, that the maintaining is not
`independent of file system for the mass storage.
`So, if we go to slide 46. Patent Owner’s primary argument in its
`Patent Owner response is that the term cache memory must have both mass
`storage and volatile memory. But even if the Board accepts this argument
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`5
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`which I'm going to address later when I discuss ground 1, the Petitioner still
`wins ground 2 given the specific art that is being combined in ground 2.
`So, if we go to slide 48, the Medin reference teaches maintaining
`networked objects by caching and mass storage. There is no dispute on this
`point in terms of the teachings of Medin and if we go to slide 49 there's the
`teachings of Markatos and Markatos teaches keeping the most frequently
`accessed documents in main memory, and that requests for documents in
`main memory are done directly without involving any file system and, again,
`there's no dispute as to this teaching either.
`So, if we go to slide 47, what we see is I've excerpted portions of our
`petition here and explanations for our arguments about the Medin and
`Markatos combination. So, it's undisputed that the combination of Medin
`and Markatos teaches a cache that includes both main memory and mass
`storage and in fact, that's the entire point of the combination that we
`provided here is a cache that can maintain network objects in both mass
`storage and in main memory or dynamic memory.
`Because specifically, if you look at the second quote that I've included
`on the bottom of side 47, Markatos teaches, and I have a quote there,
`“Keeping the most frequently accessed documents in the server's main
`memory, thus avoiding disk accesses as much as possible.”
`So, we explained a POSITA that would have applied Markatos’s
`teachings to improve Medin’s caching techniques. So, their first argument
`that the cache memory needs to include main memory and mass storage is
`exactly what the combination teaches.
`So, if we go to the second argument, their second argument on claim
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`6
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`17 is shown on -- we turn to go to slide 53. It's about the last limitation of
`claim 17 and that remaining argument is whether the combination teaches
`maintaining independent of a file system for mass storage. Well, if we go to
`slide 54 I've included an additional excerpt of Markatos. Let's keep in mind
`Markatos -- we didn’t point to Markatos for a single paragraph or, you
`know, a single sentence. Markatos’s entire -- the entire purpose of Markatos
`is to discuss this advantage of caching without having to use the file system
`and so we've highlighted the language here of one portion of Markatos, but
`it's certainly there's plenty more that we could point to. But it specifically
`states the server accesses its own cache directly without any help from the
`file system and when a client requests a document that resides in the server's
`main memory cache, the document is directly read from the server's cache
`and sent to the client without making any file system calls to access the
`document.
`So, this is crystal clear teaching of maintaining the document
`independent of the file system for mass storage and in fact, Patent Owner’s
`expert did not contest that the objects for cached -- the objects that are
`cached in main memory in Markatos are maintained independent of the file
`system for mass storage and I would point to his deposition transcript at
`pages 207 and 208 where he did not contest that issue.
`So, if we go to slide 55, what we see is that there's no dispute between
`the parties. I believe Patent Owner’s slides have an identical slide to this.
`There's no dispute between the parties that Markatos teaches maintaining 60
`to 90 percent of network objects independent of the mass storage file system.
`That's just not a fact in dispute.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`7
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`So, what is the dispute? I'm sure you're asking then, well, what is the
`dispute? The dispute goes to slide 56. Patent Owner is arguing that the
`entire cache must be maintained independent of file system and I've
`excerpted from their Patent Owner reply their argument on this point and
`done some of the highlighting here to show you that Patent Owner is arguing
`that even if one document is stored using the file system, then the claim can't
`be satisfied.
`Well, if we go to slide 57. This was the argument that was raised in
`the POPR pre-Institution and that the Board already addressed at Institution
`and the Board noted that this argument was not commensurate with the plain
`language of claim 17.
`So, if we go to slide 58, what we see is that Patent Owner now is
`arguing and attempts to distinguish its current position from what was
`already previously rejected at Institution and the Patent Owner argues that,
`well, what it's now saying is the current position is maintaining network
`objects independently from the position that was rejected in Institution by
`saying that it now argues that all objects in the cache -- rather than the entire
`cache has to be independent -- it's saying that all objects in the cache must
`be maintained independently.
`Well, there's a few problems with that argument. First off, if we go to
`slide 59, we asked the exact question to Patent Owner’s expert on this and
`we put this exact issue to the Patent Owner’s expert, whether they were
`arguing the same argument. We put the Board's decision in front of him and
`he confirmed that he’s advocating the same position that the Board already
`rejected. He said he disagreed with the Board's finding on this point and we
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`8
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`asked him, well, what evidence would you cite to the Board showing that the
`Board got this wrong, and he simply pointed back to the same claim
`language that the Board had considered with no additional evidence and
`that's on again, his deposition transcript Exhibit 1052, pages 221 through
`224.
`
`And an additional reason why, because the Board did say, hey, the
`parties are encouraged to brief this further and you know, you have an
`opportunity to discuss this issue, a further reason why Patent Owner’s
`position is wrong is shown on slide 60 and this is a simple matter of
`antecedent basis and looking at a claim language in antecedent basis. So
`you see that limitation [a] starts with -- the claim starts with receiving a set
`of networked objects in limitation [a]. [b] then states that said network
`objects are maintained. So clearly the said network objects are referring
`back to a set of network objects from limitation [a] and then claim limitation
`[c] states that the said step of maintaining so that limitation [b] or the
`previous element [b] is performed independently of the file system and that
`said step of maintaining refers to the maintaining of the said network objects
`which it is referring back to, again, the initial [a] set of network objects. So
`the claim is just requiring a set of network objects being maintained.
`But if you look at slide 61 what is clear is that Patent Owner is
`seeking to rewrite the claim and I'd note that Patent Owner did not seek to
`request here in this proceeding to amend its claims. But what it is trying to
`do in its argument is to change the language to instead of maintaining a set
`or said set of network objects, to say now that the claim should require all
`sets of network objects. But that's not how the claim was drafted and
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`9
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`written.
`So, I think slide 62, I'm sure you're familiar with this case law.
`You've seen it before. But for method claims, you're not required to show
`that it's performed all the time. You just have to show that method steps
`would be performed at least sometimes and that's clear Federal Circuit case
`law and we just provided a couple of examples here from our briefing.
`An additional reason why Patent Owner’s positions should be rejected
`is also on slide 63 and simply put, we've included on slide 63 the
`infringement contentions from the District Court and specifically, if you
`look on the left hand side, you see a limitation [17c] the limitation of what
`would satisfy wherein said step of maintaining is performed independently
`of file storage for said mass storage, sorry, for a file system for said mass
`storage.
`Now Broadcom served these contentions, Patent Owner served these
`contentions before we filed our IPR and they identified to the District Court
`what they claim satisfied this limitation [17c] and based on these
`contentions, we relied upon prior art that shows exactly the same feature that
`Broadcom argued satisfied limitation [17c].
`So, if you look in the bottom in the highlighted language that Patent
`Owner was pointing to the fact that popular files are locked into memory
`rather than fetched constantly from disk. This latter memory optimization
`eliminates the possibility that disk I/O being the cause of a server capacity
`bottleneck.
`So, what does Markatos teach? Markatos teaches us the exact same
`thing. It talks about the most popular and the most frequently accessed files
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`10
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`are cached in the main memory. So, Patent Owner has no response for
`flipping its positions from District Court and in the PTAB proceedings here.
`If you look at the sur-reply, they have no answer for the flip in positions.
`Their only statement is it's impossible to know what Broadcom was pointing
`to or was actually talking about when it provided its claim charts. That's just
`wrong. It's clear as day, it's spot on with our prior art and in fact, the ‘794
`patent itself also discusses the same functionality and what the ‘794 patent
`talks about, it describes it in terms of hot objects. But if you look at column
`8, lines 9 through 15 of the ‘794 patent itself, it talks about a subset of the
`box 200 are maintained in memory so as to use the memory 103 as a cache
`for the mass storage 104 and it talks about these hot objects being kept. So
`frequently accessed hot objects frequently in popular files, the same exact
`concept. So, Patent Owner can't run away from its positions here now to
`avoid this IPR. So that's all that I have for a claim 17 I've addressed --
`JUDGE DROESCH: Before you go, well would it be fair to say that
`the understanding of maintaining independently of the file system for the
`mass storage, would it be fair to say that that means that it's in the cache
`memory and there is no need to fetch objects from the mass storage without
`the need to fetch?
`MR. BATTS: I think that's correct. Yes, I think that's correct.
`JUDGE DROESCH: Okay. Great. Thank you.
`MR. BATTS: And I was actually about to ask before I move on from
`claim 17 are there any questions about what I've covered for claim 17?
`JUDGE DIRBA: This is Judge Dirba. I have a couple of additional
`questions, too, just to make sure that that I'm correctly understanding your
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`11
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`argument. If I understand correctly that Markatos refers -- when it refers to
`main memory, it is referring to a volatile memory like RAM; is that correct?
`MR. BATTS: Correct. There's no dispute on that point, Your Honor.
`JUDGE DIRBA: And when it is maintaining it independently, those
`objects that are being maintained independently are objects that are only
`stored in RAM and not stored in the mass storage; is that right?
`MR. BATTS: No, I don't think that's right. I think what we're saying
`is that the objects when stored in the volatile memory, so that if we go back
`to slide 55, that approximately 60 to 90 percent, or not proximate, between
`60 to 90 percent of the objects are being stored in the main memory or the
`dynamic memory in addition to the mass storage. So, when they are stored
`in the main memory, they're being stored independent of a file system of
`mass storage. That doesn't mean that a copy of that object can’t also be in
`mass storage.
`JUDGE DIRBA: I see. So, it's the storage in that main memory in
`that RAM that is satisfying this independent -- or the maintaining
`independently limitations; is that right?
`MR. BATTS: Yes, Your Honor. Well, if there's no other questions
`then I'm going to jump right back to the beginning of my slides.
`JUDGE ENGELS: Actually, before we leave this issue, could you
`address the file history in particular in Exhibit 1002? I think probably the
`most relevant pages to this issue would be pages 355 to 356.
`MR. BATTS: I don't know those pages offhand, Your Honor, so I'm
`going to turn to those. Was there a specific issue there?
`JUDGE ENGELS: Well, and if you'd like to come back to it, perhaps
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`12
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`even in your rebuttal that's okay. It addresses the language of maintaining --
`the step of maintaining is performed independently of a file system for mass
`storage and the cached memory. I'd just be interested to hear about both
`sides view of this file history.
`MR. BATTS: Okay. I will turn to that in my rebuttal. I don't recall
`that being raised before, so I'll just take a look at that and address it in my
`rebuttal. I would point out, though, that on side 55 that neither side is
`disputing that Markatos’s teachings regarding the caching of 60 to 90
`percent of the network objects in the main memory is done independent of
`the mass storage file system. So, there's simply just no dispute. The
`question is, do you have to have all the documents being or all of the
`network objects being stored independently or not, the antecedent basis issue
`that I covered. But I will address the file history on my rebuttal time.
`So then if I turn to slide 7. So now I'm turning to the arguments for
`ground 1 and Patent Owner’s main argument in its Patent Owner response is
`a claim construction issue. It's about the term cache memory and as you can
`see here I've highlighted some portions of their Patent Owner response
`where Patent Owner is contending that the term the cache memory lasting
`must include both volatile memory and mass storage.
`JUDGE ENGELS: I'm sorry, which slide?
`MR. BATTS: Slide 7, Your Honor.
`JUDGE ENGELS: Thank you.
`MR. BATTS: And I think that's a clear dispute that the parties have
`addressed in their briefing and a clear position that Patent Owner has taken
`and I've included some excerpts there, the headings and statements within
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`13
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`the POR.
`So, if we go to slide 8, I've laid out the competing proposals for
`construction of the term cache memory and if you look on the left hand side,
`Petitioner's construction is that cache memory must include mass storage but
`does not require volatile memory, whereas Patent Owner’s construction on
`the right hand side is that cache memory must include both volatile memory
`and mass storage.
`So, if we go to slide 9, I've excerpted some of the claim language to
`start with the claim language for this analysis and if you look first at the
`claim language, the claim itself states that cache memory includes mass
`storage. That's the clear, the clear language of the claim is that the cache
`memory includes mass storage. Not saying anything about volatile memory.
`It doesn't mean that it doesn't -- that it can't have volatile memory but it
`doesn't require volatile memory.
`And if we go to slide 10, I don't think there's any dispute that the
`parties agree that mass storage is nonvolatile memory. That's one of those
`things that I think is probably clear without any expert testimony and our
`arguments. But in any event, I don't think there's any dispute mass storage is
`not a volatile memory.
`So, if you go to slide 11, what we do see is that Patent Owner is
`seeking to import a limitation into the claim of volatile memory being
`required for cache memory. So, while there's no dispute that mass storage is
`a form of nonvolatile memory, so the claim is stating -- if you think about it
`-- the claim is stating cache memory includes nonvolatile memory and
`includes mass storage. But there is no similar requirement in the claim
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`14
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`language that cache memory includes volatile memory. So, the claims
`certainly do not limit cache memory to only mass storage. There could be
`other types of memory, but there's no requirement in the claim language and
`you can see with the injection of the requirement in the Patent Owner
`response where we've made in red font to include a requirement of having
`also volatile memory and we highlight that kind of the clear what Patent
`Owner is doing on slide 12. They're clearly just writing in an additional
`limitation to the claims to require that cache memory also be and also
`include volatile memory.
`But if we go to slide 14 Patent Owner said we were being
`disingenuous, that Petitioner was being disingenuous that saying that Patent
`Owner was importing a limitation into the claim and that was their statement
`in their sur-reply.
`Your Honor, I'd like to turn to slide 15.
`JUDGE ENGELS: If I may, is Patent Owner correct that every
`example in the challenged patent that uses the word memory, refers to
`volatile memory?
`MR. BATTS: I would say no, Your Honor, and I think there's a
`couple of points I want to make on that question. So, the first point would
`be, and I am going to address this as I go through, first is that the term -- you
`just used the term memory when you asked me that question, not the term
`cache memory. So, the specification of the patent does not include the term
`cache memory. So what Patent Owner is relying upon is looking at the term
`memory in the patent and I certainly don't dispute that when there is plenty
`of references to memory -- even in abstract there is a reference to memory --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`15
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`and then like, for example, even in the abstract it says memory and then it
`says (e.g., RAM). So certainly memory can include RAM or volatile
`memory, but that doesn't mean that the term memory is limited to volatile
`memory.
`So, what we saw was, and I will keep addressing this as I go through,
`I think I tried to have a logical progression on how I was going from this of
`if you look at slide 15, the Patent Owner’s expert admitted that he was
`importing into the claim limitation. We asked him specifically the claim
`language and he basically said, are you basically saying the claim language
`also has to include main memory, so volatile memory, main memory and he
`said, yes. I mean, he repeatedly said that that's what he was doing.
`So, I think slide 16 is helpful to kind of give context to this argument
`of where Patent Owner is going on this argument, which is this first step of
`their argument is that cache memory includes the term memory and then the
`second step is, well, a POSITA would have understood memory to only
`mean volatile memory and I think there's a few different problems with this,
`a few different logical leaps and problems with that sort of approach.
`Because if you look at claims, slide 17, going back to the claim
`language, the claim language alone destroys Patent Owner’s argument on
`this point because the claim language expressly states that cache memory
`includes mass storage. So already the term memory, as used in the claim, is
`referring to nonvolatile memory by saying it includes mass storage. So
`that's a point that, you know, both Petitioner’s and Patent Owner’s experts
`agreed on this point. I point to Dr. Sundaresan’s deposition 1052 at pages
`78 through 80.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`16
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`And this goes to, if you go to slide 18, this goes to the issue is that if
`you're going to go to the term memory, not cache memory, the term memory
`is not limited to volatile memory and you have to rewrite the meaning of
`memory because clearly memory has both volatile memory in it and
`nonvolatile memory as subsets of the term memory. I think that's probably
`pretty common knowledge that doesn't require a lot of expert testimony but I
`think both experts have agreed on this point and if you go to slide 19 and 20
`--
`
`JUDGE ENGELS: Just looking briefly though, again, at the claim
`language, isn't there a separate recitation of cache memory and mass
`storage? It's two separately recited elements.
`MR. BATTS: I actually -- I disagree with that, Your Honor, because
`the element of said cache memory, it says, is including mass storage. So, I
`don't view that as two separate elements that’s explaining that the cache
`memory has to include mass storage. It's a clarification of what cache
`memory has to include and so --
`JUDGE DIRBA: I have an additional question. Would you agree that
`the term memory can be used by persons of skill in the art to refer to main
`memory?
`MR. BATTS: Yes, absolutely.
`JUDGE DIRBA: Okay.
`MR. BATTS: Yes. I think the term memory, you know, the term
`memory has a lot of different usages and I think that on slide 18, you see, I
`think people can refer to RAM as you can say, memory, meaning RAM.
`You can mean DRAM. You can also say, hey, the memory on my iPhone,
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`17
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`how much memory do you have on your iPhone? I have 256 gigabytes on
`my iPhone. So, the term memory can certainly include RAM or include
`volatile memory. But our point is that it's not limited to that and --
`JUDGE DIRBA: Let me ask -- oh, I apologize for interrupting. Go
`ahead.
`MR. BATTS: No, go ahead, Your Honor.
`JUDGE DIRBA: I meant to ask a different question, so I'm glad you
`elaborated on that. My question was, do people of skill in the art or would
`you agree that people of skill in the art can use just the word memory itself
`to specifically refer to main memory as opposed to the other types of
`memory? In other words, my question is, can memory be used in two
`different ways? One is the general way that you have identified on your
`slide 18, memory can include all these different types of memory and a
`second way is sort of a shorthand that could be used specifically to refer to
`main memory.
`MR. BATTS: I think that's fair, Your Honor. I do. I think people do
`use the term memory to refer to the dynamic access memory. I mean, that's
`a RAM. But I also think as we laid out on slide 18, memory can have a lot
`of other meanings as well.
`JUDGE DIRBA: And if I understand your argument correctly, given
`the context of the claim, we know that memory has this broader meaning
`rather than the more narrow meaning because the claim says that the
`memory needs to include mass storage.
`MR. BATTS: Correct, Your Honor. That's one of my arguments.
`That's the argument I was just laying out and I'll go through a few more
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`18
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`arguments in addition to that. Any other questions before I move on?
`JUDGE DIRBA: No, thank you.
`MR. BATTS: So then slides 19 and 20 just include probably again,
`not surprising because a lot of us have seen memory in a lot of patents
`probably, but there's a lot of different usages for a memory. You can talk
`about flash memory and volatile memory, nonvolatile memory. So we just
`included that Patent Owner’s expert certainly acknowledged that there's lots
`of different types of memory in slides 19 and 20 including nonvolatile and
`volatile memory.
`So, then that turns us to the dictionary definitions. So, if we go to
`slide 21 we have the various dictionary definitions. Again, these dictionary
`definitions are for memory, not for cache memory, but dictionaries -- the
`point that I would have is that dictionaries clearly don't support narrowing
`the term memory to volatile memory, let alone the term cache memory. So,
`if you look at the dictionary definitions, memory is a place where
`information is stored and it can have a lot of different meanings on that. But
`it certainly -- these dictionaries don't support a narrowing of the term of
`cache memory.
`JUDGE ENGELS: So, help me understand this. At least the way I
`understood you to say there, I think -- well, let's take as given that memory
`is a broad term. But were you to argue that cache memory is more broad
`than memory? I don't -- I'm not sure your point there.
`MR. BATTS: No, my point -- I guess I'll say it another way and
`maybe an easier way to understand, the tension point I feel like on a lot of
`Patent Owner’s arguments is that they are trying to, they take the second
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`
`
`
`
`19
`
`
`
`IPR2021-01319
`Patent 7,103,794 B2
`logic leap of trying to say, if you look at the term memory, that gives you a
`basis to construe the term cache memory and so I view the dictionaries as
`they clearly if they want to have a narrowing of the term cache memory and
`they want to say cache memory has to mean because it has memory in the
`term cache memory, you have to look at the word memory and memory
`must be dynamic memory. There has to be volatile memory and my point is,
`is even if you look at these dictionaries that are submitted as evidence in the
`record that simply meet -- the dictionary definitions don't support that logic
`because the term memory, even if you look here when it gives the examples
`on the Microsoft Press dictionary, is talking about EEPROM, but it's also
`talking about flash memory. It's talking about RAM. It’s talking about
`ROM. It's clearly talking abou