`
`Trials@uspto.gov
`571-272-7822 Date: February 4, 2015
`
`
`RECORD OF ORAL HEARING
`UNITED STATES PATENT AND TRADEMARK OFFICE
`- - - - - -
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`- - - - -
`DELL INC., HEWLETT-PACKARD COMPANY and NETAPP, INC.
`Petitioners
`vs.
`ELECTRONICS AND TELECOMMUNICATIONS RESEARCH
`Patent Owner
`- - - - -
`Appeal No. IPR2013-00635
`Patent No. 6,978,346 B2
`- - - - - -
`Oral Hearing Held: December 18, 2014
`
`
`Before: BRIAN J. McNAMARA, MIRIAM L. QUINN (via video),
`
`GREGG I. ANDERSON (via video), Administrative Patent Judges.
`
`The above-entitled matter came on for hearing on Thursday,
`
`December 18, 2014 at the U.S. Patent and Trademark Office, 600 Dulany
`Street, Alexandria, Virginia at 1:00 p.m., in Courtroom A.
`
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONERS:
`
`
`JAMES RUSSELL EMERSON, ESQ.
`
`
`DAVID L. McCOMBS, ESQ.
`
`
`Haynes and Boone LLP
`
`
`2323 Victory Avenue, Suite 700
`
`
`Dallas, Texas 75219
`
`
`214-651-5533
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`THOMAS KELTON, ESQ.
`Hanes and Boone LLP
`2505 N. Plano Road, Suite 4000
`Richardson, Texas 72082-4101
`
`ON BEHALF OF THE PATENT OWNER:
`
`DEREK MEEKER, ESQ.
`
`MATTHEW C. PHILLIPS, ESQ.
`
`Renaissance IP Law Group LLP
`
`9600 SW Oak street, Suite 560
`
`Portland, OR 97223
`
`503-419-6425
`
`
`
`
`
`
`
`
`
`ALEXANDER GIZA, ESQ.
`Russ August & Kabat
`12424 Wilshire Boulevard, 12th Floor
`Los Angeles, CA 90025
`
`
`
`2
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`P R O C E E D I N G S
`(1:00 p.m.)
`
`
`
`
`
`
`
`JUDGE McNAMARA: Please be seated.
`Good afternoon. This is the oral hearing in case
`IPR-2013-00635. Petitioners are Dell, Inc., Hewlett-Packard,
`Company, and NetApp, Inc. And the Patent Owner is
`Electronics and Telecommunications Research Institute.
`I am Judge McNamara. Judges Quinn and
`Judge Anderson are participating remotely, so you can see
`them on the screen.
`And could I have the Petitioners' counsel please
`introduce yourselves.
`MR. McCOMBS: Yes, Your Honors. This is
`David McCombs for the Petitioners, Dell, Hewlett-Packard,
`and Net App. With me are counsels Russ Emerson and Thomas
`Kelton. Russ Emerson will be making the presentation today.
`And also with us are client representatives Ann Byun for
`Hewlett-Packard, and Cynthia Rosser for NetApp.
`JUDGE McNAMARA: Thank you. And Patent
`
`Owner?
`
`MR. PHILLIPS: Good afternoon, Your Honor.
`Matthew Philips for the Patent Owner. With me are my
`co-counsel, Alexander Giza and Derek Meeker.
`JUDGE McNAMARA: Thank you. Welcome to the
`Patent Trial and Appeal Board. We have no motions pending
`
`
`
`3
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`before us today, other than the case concerning the petition
`itself.
`
`Each party will have 45 minutes. We will begin
`with the Petitioner, who bears the ultimate burden of proof and
`may reserve some time to rebut the opposition put on by the
`Patent Owner.
`Following the Petitioner's initial argument, the
`Patent Owner may use up to -- will have up to 45 minutes to
`oppose, and then the Petitioner can use any time if reserved for
`rebuttal.
`
`Is everybody ready? Let's begin.
`MR. EMERSON: Yes, Your Honor. Before I
`begin, would you like a hard copy of our demonstratives?
`JUDGE McNAMARA: Sure. Thank you very
`much. And is there some amount of time you would like to
`reserve for rebuttal?
`MR. EMERSON: I would like to reserve 15
`minutes for rebuttal, Your Honor.
`JUDGE McNAMARA: Okay. So I will let you
`know when 30 minutes is up.
`MR. EMERSON: Thank you very much, Your
`
`Honor.
`
`And the judges who are joining us remotely, thank
`you, and good afternoon. My name, again, is Russ Emerson for
`the Petitioners.
`
`
`
`4
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`We will just dive into this, going first to slide 2.
`The '346 patent has one independent claim that is at issue in
`this IPR. That is independent claim 1.
`Independent claim 1 recites four basic or
`fundamental elements, a RAID, two RAID controllers, each
`RAID controller having at least two network controlling units,
`and a plurality of connection units.
`Now, importantly, claim 1 doesn't claim any
`relationship between the RAID of the preamble and the
`controlling units, nor between the controlling units and the
`disk drives that make up the RAID. As we have shown in our
`briefing, Your Honors, Hathorn teaches all of these elements.
`Moving on to slide number 3, this is figure 3 from
`Hathorn, and Hathorn discloses a RAID and RAID controlling
`units. We see the RAIDs -- rather the DASD 326 and 336 that
`make up the RAID, the redundant array of independent disks,
`storage controllers 325 --
`JUDGE McNAMARA: Excuse me, you used the
`term DASD. That's another acronym. Could you at least just
`for the record tell us what it is?
`MR. EMERSON: For the record, a DASD, as
`explained in Hathorn, stands for direct access storage device.
`JUDGE McNAMARA: That's spelled D-A-S-D?
`MR. EMERSON: Yes, sir. Back to figure 3,
`storage controllers 325 and 335 are the RAID controlling units.
`
`
`
`5
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`The ports A through D on each of those storage controllers are
`the network control units. And dynamic switches 315 and 305
`are the connecting units.
`And what Hathorn describes is a shadowing or a
`mirroring operation that works as follows: The primary host
`writes to the first storage controller 326 -- I'm sorry, 325
`through dynamic switch 305. That storage controller writes the
`data to be mirrored or shadowed to DASD 336 or, rather,
`DASD 325.
`JUDGE McNAMARA: I'm sorry, is it DASD 325
`or 326? I think it is storage controller 325 and DASD 326, is
`what I think you have there.
`MR. EMERSON: Yes.
`JUDGE McNAMARA: Okay.
`MR. McNAMARA: Yes, that's right, Your Honor.
`Controller 325 then sends that data to be shadowed
`to controller 335. And controller 335 saves that data to DASD
`336. And then storage controller 335 tells storage controller
`325 that the data was successfully mirrored or shadowed on the
`second DASD.
`Moving on to slide 4, we have the Court's -- or the
`Board's construction of RAID. And the RAID is construed by
`the Board simply as its acronym. It's a redundant array of
`inexpensive disks. And the Board has consistently construed
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`6
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`RAID to mean just what it says throughout all of the IPRs
`regarding this patent.
`The Patent Owner asks the Board to construe RAID
`to be a lot narrower than its actual -- than what the words of
`the acronym actually mean. They want the RAID to mean a
`single logical unit for mass storage that provides fault
`tolerance and recovery via employing multiple physical disk
`drives, which is a lot to pack into the acronym RAID, which
`we respectfully submit the Board got correct when it construed
`to mean exactly what it says, a redundant array of inexpensive
`disks.
`
`The Patent Owner's proposed construction packs
`into a RAID a number of unclaimed and extraneous limitations
`that make it far more narrow than the broadest reasonable
`interpretation.
`Now, moving on to slide 5, the Patent Owner has
`contended that a RAID must be a single logical unit. But a
`RAID need not be limited to a single logical unit. We show
`here an excerpt from the Weygant reference that states that a
`RAID can be configured in many ways, either as a single unit
`or in various combinations of striped and mirrored
`configurations. So Weygant tells us that a RAID need not be
`configured as a single logical unit. So a RAID can mean many
`things. It means different things to different people, but the
`broadest reasonable interpretation, we respectfully submit,
`
`
`
`7
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`would exclude this limitation, this extraneous limitation of a
`single logical unit.
`Now, moving on to slide 6, Hathorn teaches a
`RAID because Hathorn shows an array of disk drives set up to
`do mirroring or shadowing. And Hathorn certainly discloses
`shadowing. That's the title of the patent, remote data
`shadowing.
`Moving on to slide 7, this is a longer excerpt from
`the Weygant reference. And you can see at -- RAID Level 1 is
`defined as disk mirroring. And, actually, if you look at the
`Hathorn patent, Hathorn describes dual storage both as
`mirroring and shadowing.
`So Hathorn teaches shadowing. It discusses dual
`storage as shadowing or as mirroring. RAID includes
`mirroring. And mirroring and shadowing are both two -- those
`are two words to describe the same concept in RAID. And here
`on slide 8 is an excerpt from Dr. Mercer's declaration, which
`cites the Chen reference, which is before the Board, and Chen
`says the traditional solution called mirroring or shadowing
`uses twice as many disks as a non-redundant array.
`So shadowing is mirroring. Mirroring is RAID,
`Number 1. Hathorn teaches shadowing. Therefore, Hathorn is
`a RAID.
`
`Moving on to slide 9, slide 9 is an excerpt from
`column 2 of Hathorn.
`
`
`
`8
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`JUDGE QUINN: Counsel, I have a question.
`MR. EMERSON: Yes, Your Honor.
`JUDGE QUINN: So you are urging this panel to
`not alter the claim construction we have already provided that
`says RAID means a redundant array of inexpensive disks and
`not to add the other things as fault tolerance and things of the
`like, but yet you are arguing that Hathorn, because it does
`perform those additional functions of a RAID, that it does
`disclose a RAID. So my question to you is since there is an
`argument about whether the Hathorn DASDs may perform as a
`RAID, does that mean that we need to construe a RAID to
`include these in the performance of those functions that are in
`this paper?
`MR. EMERSON: I don't believe so. What we're
`saying is that the concept of RAID encompasses a number of
`different functionalities. One of those is mirroring. And
`because Hathorn teaches mirroring or shadowing, that it
`necessarily teaches a RAID.
`Secondarily, or perhaps primarily, it teaches this
`redundant array of disks, you have two disks that are arranged
`together to provide a backup of the data and the shadowing
`configuration, so even leaving aside the definition of RAID
`Level 1 as mirroring, you can get there simply by looking at
`figure 3, which shows a redundant array of independent disks.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`9
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`JUDGE QUINN: Which two disks in figure 3 are
`you contending are an array?
`MR. EMERSON: The two disks in the two DASDs
`in figure 3 are 326 and 336. You can see it on slide 3, where
`we've got it highlighted. DASDs 326 and 336 are the RAID.
`Those are the DASDs, the disks that make up this RAID.
`JUDGE ANDERSON: So, counsel, does a single
`DASD have a single disk?
`MR. EMERSON: Hathorn is unclear about that,
`Your Honor. So it could be a single disk. It could be multiple
`disks.
`
`JUDGE QUINN: And --
`JUDGE ANDERSON: Does Hathorn say anything
`specific about that?
`MR. EMERSON: I don't believe so, Your Honor.
`JUDGE QUINN: What is your explanation for
`DASD 323 including multiple volumes?
`MR. EMERSON: We're not quite sure. That is --
`you see that 323 shows volume A and B. The second DASD,
`326, doesn't say anything other than DASD. And then you've
`got A primed and B primed. We're not quite sure what that
`means necessarily. We think it is some sort of a typo.
`JUDGE QUINN: Does your expert have any
`opinions as to what that diagram teaches one of skill in the art
`regarding volume changes?
`
`
`
`10
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`MR. EMERSON: Specifically for DASD 323?
`JUDGE QUINN: Yes.
`MR. EMERSON: I don't believe so, Your Honor.
`And since we're looking at DASDs 326 and 336,
`that's not that germane to our argument. And we -- and for the
`Board's benefit, we're talking about those DASDs because
`those are the DASDs that Hathorn walks through to explain its
`mirroring operation.
`Okay. So moving on back to slide 9, this is the
`excerpt from the background of the invention that we cited in
`our petition. It was given a lot of attention by the Patent
`Owner in its responsive briefing. And so I want to address that
`briefly for the Board.
`We cited this -- and, for the record, this citation is
`to the first part of column 2 at lines 4 through 11 of the
`Hathorn reference, and it talks about DASDs being set up in a
`RAID configuration. And the only reason we cited that, Your
`Honors, is for the simple proposition that we think is
`unremarkable that DASDs can be arranged in a RAID
`configuration. That's it.
`We think figure 3 stands on its own. We think
`figure 3 has to stand or fall on its own. If figure 3 didn't show
`a RAID configuration, if it didn't show mirroring, then this
`citation to the background of the invention wouldn't save us.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`11
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`And since it does, this citation simply supports the idea that
`DASDs can be arranged as a RAID.
`So that's point number 1. We don't tie this to
`figure 3. We don't say that this citation means that figure 3 is
`a RAID. We cite it simply for the unremarkable proposition
`that DASDs can be arranged as a RAID.
`The second --
`JUDGE QUINN: When you say it is unremarkable,
`because -- because the Patent Owner is not thinking it is
`unremarkable, so we want to know what is remarkable about
`that statement.
`MR. EMERSON: Well, we think it is not that
`remarkable, that DASDs can be arranged as a RAID. DASDs
`are simply storage devices. Disks fall within the concept of
`DASDs. And a RAID is redundant array of independent disks.
`So when you have an array of disks and DASDs
`include disks, it makes sense that DASDs can be arranged in a
`RAID configuration.
`Does that answer your question, Your Honor?
`JUDGE QUINN: It will have to do for now.
`MR. EMERSON: Okay. The other point about this
`is that this discussion here distinguishes what they call a RAID
`from dual storage, which they discuss in the previous
`paragraph.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`12
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`Dual storage is mirroring. That's what Hathorn
`calls it. So what they are doing here is they are distinguishing
`a type of RAID from mirroring. The type of RAID they are
`talking about here is an apportionment of data amongst
`DASDs, which is not mirroring. So they are distinguishing
`what we think is some sort of a striping-type RAID from a
`mirroring RAID. So that's what they are doing.
`So just to go back and summarize, this citation
`doesn't make figure 3 a RAID; figure 3 stands on its own as a
`RAID. And this citation distinguishing mirroring from a
`non-mirroring RAID doesn't affect that analysis. Mirroring is
`RAID, whether Hathorn calls it that or not.
`JUDGE ANDERSON: Do you mean it has to be a
`RAID if there is mirroring going on?
`MR. EMERSON: Your Honor, mirroring -- RAID
`Level 1 is defined as mirroring. What they are doing there in
`figure 3 is mirroring. So, therefore, it is our contention that
`figure 3 shows mirroring and it shows RAID. We haven't seen
`anything out there in the record that distinguishes some sort of
`a non-RAID mirroring from a RAID mirroring. There is
`nothing in the record to support that. So the answer is yes,
`Your Honor.
`JUDGE McNAMARA: Counsel, you just have --
`maybe I am not reading Hathorn right, but you just referred to
`the paragraph before the one that you have on slide 9 and said
`
`
`
`13
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`that the paragraph on slide -- that you have on slide 9 shows
`that DASDs can be configured as a RAID.
`The interesting thing about that -- and the
`paragraph before that talked about mirroring. And so I am
`looking at that paragraph, the first paragraph, the one that is
`not shown on the slide, and it says an example of dual copy
`involves providing additional DASDs so that data is written to
`the additional DASDs, sometimes called mirroring.
`Then the paragraph you have there says another
`alternative overcomes the need to double the storage devices,
`the data -- and then refers to the RAID. And then it says, "The
`data is written such that the data is apportioned among the
`DASDs. So if a single DASD fails, then the lost data can be
`recovered by using the remaining data and error correction
`procedures."
`So I read that paragraph that you have up there --
`the rest of the paragraph that you have up there on slide 9 to
`say something much different. It is not mirroring, because you
`are going to error correction procedures to get the data back.
`You are not necessarily going to store all the data. You are
`just going to store enough to recover -- to get a recovery.
`MR. EMERSON: That's absolutely correct, Your
`Honor. We agree. That's the point I am trying to make.
`What they are talking about there in column 2 is an
`apportioning-type RAID, which is likely RAID 5 or something
`
`
`
`14
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`like that because it talks about error correction, okay? And
`they are distinguishing that from mirroring, which is
`discussed -- which is discussed below at the bottom of column
`1.
`
`JUDGE McNAMARA: Okay.
`MR. EMERSON: So they are saying apportioning
`RAID is different from mirroring because you don't have to
`double the disks. But in mirroring, you have to double the
`disks. It is double the disk drives. So that that's what they are
`distinguishing. That is exactly the point we're making, Your
`Honor.
`
`JUDGE McNAMARA: All right.
`MR. EMERSON: Moving on to the next slide,
`which is slide 10, this whole discussion of RAID, we think,
`is -- you can summarize it by noting the Patent Owner's expert,
`Dr. Conte, who admits in his declaration that mirroring is
`provided by figure 3, and his only dispute with that is whether
`or not there is one storage controller or two.
`Rather, one RAID controlling unit or two.
`So moving on to slide 11, this is the Board's
`construction of RAID controlling unit, which the Board
`construed to mean effectively what it says: A RAID
`controlling unit or RAID controller is a component that
`controls operation of the RAID.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`15
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`And here is the Patent Owner's construction of
`RAID controlling unit. And we respectfully submit this is far
`too narrow. They are talking about the RAID controlling unit
`being a component that controls operation of the RAID, so far
`so good, so as to provide redundant storage of data among the
`array of disk drives. For our purposes, that might be okay, but
`that would exclude striping or apportionment-type RAID. But
`the real problem here is the corollary. It says, "A corollary of
`this interpretation is that the RAID controlling unit must be
`able to write to and read from all the disk drives of the RAID
`array."
`
`There is nothing in the intrinsic evidence that talks
`about the relationship between the RAID controlling units and
`disk drives of the RAID array. If you look at figure 4 of the
`'346 patent, it doesn't show the disk drives at all. It doesn't
`show any relationship between the disk drives and the RAID
`controllers.
`So this is something that is entirely extrinsic to the
`record here. The only support for it is Dr. Conte's unsupported
`declaration.
`So moving on to slide 13, and slide -- and let me
`move back.
`Hathorn teaches the RAID controllers because
`Hathorn teaches the storage controllers that control the
`mirroring operation of Hathorn. So this is back to slide 10.
`
`
`
`16
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`The mirroring operation is controlled by storage controllers
`325 and 335. So that's why Hathorn teaches RAID controllers.
`Next is slide 13. Slide 13, again, is figure 3. We
`have the first and second network controlling units and the
`third and fourth network controlling units. Those are the
`network controlling units. Storage controller 325 has these
`ports A through D. All of them have these ports A through D.
`And those are the controlling units in Hathorn.
`And those are the controlling units in Hathorn because those
`ports allow the storage controllers or the RAID controllers to
`communicate over the network.
`The Board did not construe this term. On slide 14,
`the Board did not construe this term but gave it its plain and
`ordinary meaning, which is consistent with the construction
`that the Petitioners urged, which was simply that it is a
`component that allows the controllers to communicate over the
`network.
`
`JUDGE McNAMARA: Could I ask you just quickly
`to go back to slide 13 for just a second?
`MR. EMERSON: Yes, Your Honor.
`JUDGE McNAMARA: You have identified those
`things that is the first and second network controlling units.
`MR. EMERSON: That's right.
`JUDGE McNAMARA: And claim 1, I think, says,
`if I read it right, that the first network controlling unit is
`
`
`
`17
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`coupled or connected directly to the fourth network controlling
`unit. So exchanges information with the fourth network
`controlling unit, and the second network controlling unit
`exchanges information with the third network controlling unit.
`MR. EMERSON: That's right.
`JUDGE McNAMARA: Could you just demonstrate
`to me how that happens in that slide?
`MR. EMERSON: Sure. So we have -- the first rate
`controlling unit --
`JUDGE McNAMARA: We have to be careful about
`the terminology here, because we get RAID control unit and
`network control unit.
`MR. EMERSON: Right.
`JUDGE McNAMARA: And these are not net RAID
`control units; these are the network controlling units.
`MR. EMERSON: Right. The first network
`controlling unit exchanges information with the fourth network
`controlling unit.
`JUDGE McNAMARA: Right.
`MR. EMERSON: So we had the first network
`controlling unit, which is port A in 324. And it connects
`through the dynamic switches down to the fourth network
`controlling unit, which you could call either A or B in storage
`controller 336.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`18
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`JUDGE McNAMARA: Okay. So it is through the
`dynamic switches?
`MR. EMERSON: It is through the dynamic
`
`switches.
`
`JUDGE McNAMARA: Okay. Now, the first unit,
`okay -- and this is where -- then the first RAID controlling unit
`and the second RAID controlling unit, so those are the RAID
`controlling units as opposed to the network control units; is
`that right?
`MR. EMERSON: That's right.
`JUDGE McNAMARA: Do you agree with that part
`of the claim? Okay. Because it says they exchange
`information, the second RAID controlling unit, directly
`exchange information with the host computers. Okay.
`How does that happen? And where is that
`happening there? Is that happening in the switch as well?
`MR. EMERSON: They exchange information with
`the host computers through the dynamic switches. And that's
`what the claim language says. They exchange information with
`the numerous host computers through the plurality of
`connecting units.
`JUDGE McNAMARA: -- through the plurality of
`connecting units.
`MR. EMERSON: And in Hathorn --
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`19
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`
`JUDGE McNAMARA: And the dynamic switch is
`the connecting unit?
`MR. EMERSON: Yes, Your Honor.
`JUDGE McNAMARA: Okay. All right. I just
`wanted to make sure I had the mapping.
`MR. EMERSON: Right. And that's an important
`point throughout, that connecting includes connecting through
`these switches or hubs.
`The Patent Owner's argument here is that it could
`be that these ports in Hathorn are simply a single network
`interface card or network interface adapter that has multiple
`ports in it. And in Dr. Conte's declaration, he has pictures
`here shown on slide 15 of some network cards, saying it is
`possible that the ports are simply the ports of network cards
`and, therefore, there aren't multiple network controlling units
`involved here.
`And I would say, first of all, to that, that the
`Hathorn patent specifically talks about these ports as being
`separate and distinct components. It talks about them
`controlling the communication between the storage
`controllers -- between the storage controllers themselves and
`between the storage controllers and the hosts.
`And the patentee did not use that term in the
`claims. And we all know what a NIC is, what a network
`interface card is, what an adapter is. Those terms connote
`
`
`
`20
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`specific hardware that the patentee could have chosen to use,
`but he didn't. He used plain, generic language.
`And here, on claim 16, we have a couple of
`definitions that we had in our reply. Looking first at the IEEE
`standards, defines network interface controller simply to be a
`communication device which permits the connection of
`information processing devices to a network, which I think is
`consistent with the Board's plain and ordinary construction as
`well as our proposed construction.
`And you look below that at the definition of
`network interface card, and that connotes certain specific
`hardware. We talk about a NIC, we talk about an adapter. The
`patentee did not use those terms.
`Moving on to slide 17, this is just from
`Dr. Mercer's declaration, making the same point.
`Slide 18 we have a citation from Hathorn at column
`1, where they talk about these ports are able to be dynamically
`set and communicate either as a channel or control unit link
`level facility. So these are separate and distinct
`communication units that are -- that are separately called out in
`Hathorn, and it wouldn't be right to say, well, it could be that
`those are all just single physical sockets in a NIC.
`So that's claim 1. We've got a RAID because it is
`mirroring, we've got RAID controllers that control the
`mirroring operation. We have two of those. And we have the
`
`
`
`21
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`network interface controllers, the network controlling units
`that are the ports. And we also have --
`JUDGE QUINN: Counsel, I have a question.
`MR. EMERSON: Yes, Your Honor.
`JUDGE QUINN: If I may. Going back to the
`definition of RAID, Patent Owner has provided some extrinsic
`evidence of what the acronym means to a person of ordinary
`skill in the art. Some of that evidence appears to focus on the
`fact that these various computer disks behave or act as a single
`storage unit.
`If we were to adopt such an interpretation of the
`acronym, how do the DASDs in Hathorn behave as a single
`storage unit?
`MR. EMERSON: If you look at column -- I believe
`it is column 9, where there is a discussion of the mirroring of
`the shadowing operation, you can see that once the data is
`written from the primary host to the storage controller through
`the dynamic switch, the primary host has nothing more to do
`with that data. The entirety of the mirroring operation is
`carried out by the storage controllers, not by the host.
`So we would submit -- and we would not concede
`that Hathorn shows a RAID that has multiple disks that are --
`that are seen by the host as a single logical unit. After
`initialization, which is described in column 8, the mirroring
`process is carried out entirely by the storage controllers and
`
`
`
`22
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`not by the host. So we would say even if you did adopt their
`construction, which we don't think would be proper -- we think
`that is more narrow than the broadest reasonable interpretation,
`and we have also given the Board some extrinsic evidence that
`indicates that RAIDs can be an array of disks that either are or
`are not set up as a single logical unit. But if you were to go
`that way, Hathorn still anticipates.
`Let me run through the rest of my slides quickly. I
`know I am over my 30 minutes.
`JUDGE McNAMARA: Yeah, you are about 13 --
`you've got about 13 and a half minutes left.
`MR. EMERSON: Okay. Moving on to claim 2 --
`and I will address these collectively. Now I am on slide 19.
`Claims 2, 3, and 8, the Patent Owner claims, do not -- are not
`anticipated by Hathorn because of this whole connection term.
`They are talking about, in those dependent claims, connecting
`or coupling components.
`And as we have discussed, these connecting terms
`or coupling terms include connecting through the connecting
`units, connecting through hubs or switches. That's shown in --
`it is shown in the '346 patent; it is shown also in the claims.
`So here we have from -- and I have moved on to
`slide 23, where the specification talks about the RAID
`controllers being connected with one another through the hubs.
`And we have in claim 1, as we discussed earlier,
`
`
`
`23
`
`
`
`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
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`Judge McNamara, the connecting units for connecting the
`RAID controlling units to the host computers.
`So the idea of connecting in the '346 patent clearly
`encompasses connections that go through the hubs and
`switches. So that addresses -- that is their only argument on 2,
`3, and 8. Actually, there is one other argument.
`They say that we construed "connected" -- we
`construed "coupled to" to include indirect connections. And
`that, therefore, "connected" must mean something more narrow
`than that. And I am here on slide 24 now.
`We have construed only "coupled." "Coupled"
`shows up only in some of the dependent claims. It is not in the
`specification. "Connected" clearly means connecting through
`the hubs and switches. We have shown that in the spec. We
`have shown that in the claims.
`We construed "coupled" to make it clear that
`coupled is not something that is more narrow than "connected."
`So we didn't construe "coupled" and we only construed -- we
`didn't construe "connected." We only construed "coupled" to
`make clear that that wasn't something narrower than
`"connected."
`Moving on to slide 26, now, the final three claims
`at issue are claims 5, 6, and 7. The first argument the Patent
`Owners have is that Hathorn doesn't teach hubs because they
`show these dynamic switches. But the Patent Owner or the
`
`
`
`24
`
`
`
`IPR2013-00635
`Patent 6,978,346 B2
`
`patentee, when he drafted the claims, defined "hub" to mean a
`hub or a switch. And here at slide 27, I have a citation from
`the '346 patent at column 3, lines 13 to 18.
`And here the patentee clearly, expressly, and
`explicitly acted as his own lexicographer and defined "hu