` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`Page 1
`
` RIVERBED TECHNOLOGY, INC.; DELL INC.; )
` HEWLETT-PACKARD ENTERPRISE CO.; HP
` ) CASE IPR2016-00972
` ENTERPRISE SERVICES, LLC; TERADATA
` ) PATENT 7,415,530
` OPERATIONS, INC.; ECHOSTAR
` )
` CORPORATION; AND HUGHES NETWORK
` ) CASE IPR2016-01002
` SYSTEMS, LLC,
` ) PATENT 9,116,908
` PETITIONERS
` )
` )
` )
` )
` )
` )
`
` REALTIME DATA LLC,
` PATENT OWNER
`
` VS
`
` --------------------------------------------
`
` ORAL DEPOSITION OF
`
` CHARLES D. CREUSERE, PH.D.
`
` January 19, 2017
`
` --------------------------------------------
`
`Reported by:
`
`Ronald R. Cope
`
`Job no: 17930
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`Patent Owner Ex. 2002
`Teradata Operations v. Realtime
`IPR2017-00806
`
`
`
`Page 2
`
`Page 4
`
` A P P E A R A N C E S (Continued)
`
` FOR THE PATENT OWNER:
` Mr. Kayvan B. Noroozi, Esq.
` NOROOZI, PC
` 1299 Ocean Avenue
` Suite 450
` Santa Monica, California 90401
` 310.975.7074
` e-mail: kayvan@noroozipc.com
`-and-
` Mr. Jason D. Eisenberg, Esq.
` Mr. Jay L. Bird, Esq.
` STERNE, KESSLER, GOLDSTEIN, FOX
` 1100 New York Avenue, NW
` Washington, DC 20005
` 202.371.2600
` e-mail: jasone@skgf.com
` jbird@skgf.com
`
`Page 5
`
` I N D E X
`Appearances
`CHARLES D. CREUSERE, PH.D.
` EXAMINATION BY MR. NOROOZI
` 163
`Changes and Signature
`Reporter's Certificate
` 165
`
` 3
`
` 6
`
` EXHIBITS REFERRED TO
` FROM PREVIOUS DEPOSITION(S)
` PAGE
`NUMBER
`
`Exhibit Declaration of Charles D. 7
` Creusere, Ph.D. (exhibit number
` not given)
`
`Exhibit 1001 United States Patent 7,415,530 34
`
`Exhibit 1002 Declaration of Charles D.
` Creusere, Ph.D.
`Exhibit 1005 United States Patent 5,247,646 108
`Exhibit 1011 Claim 24 147
`
` 10
`
`1
`
`23
`
`4
`
`5
`
`6
`
`7
`8
`9
`
`10
`
`11
`
`12
`
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
`1
`2
`3
`4
`5
`
`678
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` ORAL DEPOSITION OF CHARLES D. CREUSERE,
` PH.D., produced as a witness at the instance of the
` Patent Owner, and duly sworn, was taken in the
` above-styled and -numbered cause on January 19, 2017,
` from 9:03 a.m. to 2:30 p.m., before Ronald R. Cope, a
` CSR in and for the State of Texas, Registered
` Professional Reporter and Certified Realtime Reporter,
` reported by machine shorthand at the Renaissance Hotel,
` 900 E. Lookout Drive, Richardson, Texas 75082, pursuant
` to Patent Owner Realtime Data LLC's Notice of Deposition
` of Charles D. Creusere, Ph.D., and the provisions stated
` on the record.
`
`Page 3
`
` A P P E A R A N C E S
`
` FOR THE PETITIONER:
` Mr. John Russell Emerson, Esq.
` HAYNES AND BOONE, LLP
` 2323 Victory Avenue
` Suite 700
` Dallas, Texas 75219
` 214.651.5328
` e-mail: russ.emerson@haynesboone.com
` -and-
` Mr. Kyle L. Howard, Esq.
` Mr. Gregory P. Webb, Esq.
` HAYNES AND BOONE, LLP
` 2505 N. Plano Road
` Suite 4000
` Richardson, Texas 75082-4101
` 972.739.6931
` e-mail: kyle.howard@haynesboone.com
` greg.webb@haynesboone.com
` -and-
` Mr. Andrew R. Sommer, Esq.
` WINSTON & STRAWN, LLP
` 1700 K Street, NW
` Washington, DC 20006
` 202.282.5896
` e-mail: asommer@winston.com
`
` -and-
`
` Mr. Andrew D. Wilson, Esq.
` BAKER BOTTS, LLP
` The Warner
` 1299 Pennsylvania Avenue, NW
` Washington, DC 20004-2400
` 202.639.1312
` e-mail: andrew.wilson@bakerbotts.com
`
`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
`
`1
`
`23
`
`4
`
`5
`
`6
`
`7
`8
`9
`
`10
`
`11
`
`12
`
`13
`14
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`24
`25
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`2 (Pages 2 to 5)
`
`
`
`Page 6
`
`Page 8
`
`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
`
`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
`
` CHARLES D. CREUSERE, PH.D.,
`having been first duly sworn, testified as follows:
` EXAMINATION
`BY MR. NOROOZI:
` Q. Good morning, sir.
` A. Good morning.
` Q. You're here to testify today as to both the
`'530 and '908 patents, correct?
` A. Yes.
` Q. And specifically with respect to opinions that
`you've set forth in your declarations in IPR proceedings
`as to those patents, right?
` A. That is correct, yes.
` Q. Now, both of the Claim 1s of the '530 and '908
`patents require a data accelerator, true?
` A. That is correct, yes.
` Q. And the data accelerator in both of the
`Claim 1s of the '530 and '908 patents has to compress at
`least two data blocks, right?
` A. I believe so. Let me just take a quick look at
`that.
` Q. And for the record, you're taking a look --
` A. Yes.
` Q. -- at one of your declarations?
` A. Yes. The declaration for the '530 patent.
`
`Page 7
`
` MR. NOROOZI: And so we'll mark that as an
` exhibit with the same exhibit number as it has in the
` proceeding.
` A. And your question -- could you repeat the
` question again with respect to Claim 1?
` Q. (BY MR. NOROOZI) With respect to Claim 1 of
` both of the '530 and '908 patents, the data accelerator
` has to compress at least two data blocks, right?
` A. Yes. It says specifically, "Said data stream
` includes a first data block and a second data block."
` Q. Okay. Now, the two data blocks in both
` Claim 1s of both patents must be compressed using two
` different compression techniques, right?
` A. That is my understanding of the -- of both the
` '530 patent and the '908 patent.
` Q. And when we talk about "compression
` techniques," that's the same thing as compression
` algorithms, right?
` A. Yes. Compression techniques, in my
` understanding, is the same -- is synonymous with
` compression algorithms.
` Q. The data accelerator in both Claim 1s of the
` '530 and '908 patents must also compress and store the
` two data blocks faster than those same two data blocks
` would be stored in received or uncompressed form, right?
`
`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
`
`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
`
` MR. SOMMER: Object to form.
` A. Okay. I'm going to -- just to compare, I'm
` going to look at my deposition (sic) from the '908
` patent as well because the language is a little bit
` different in those two, in the claims in the two
` patents.
` Q. (BY MR. NOROOZI) And I just want to clarify
` for purposes of the deposition: Do you understand that
` you're not supposed to take any cues or draw any hints
` as to how you should answer my questions based on
` whether or not your counsel objects?
` A. Yes. Yes. No. Yes, I understand that.
` Q. Okay. So --
` A. Okay.
` Q. -- why don't we read back my question, and then
` we can take the answer from there.
` Well, I just I want to make sure it's
` going to go on the record again, so let me just put it
` on the record again. Withdrawn.
` The data accelerator in both Claim 1s of
` the '530 and '908 patents must also compress and store
` the two data blocks faster than those same two data
` blocks would be stored and received in uncompressed
` form, right?
` MR. SOMMER: Object to form.
`
`Page 9
`
` A. That -- that is my interpretation. The claims
` language in the two Claim 1s are a little bit different.
` In the Claim 1 for the '530 patent, it specifically
` says -- let's see -- a data stream is received by the
` said data accelerator in received form. The said data
` stream includes a first and second block. And then it
` says the said data stream is compressed by the data
` accelerator to provide a compressed stream by
` compressing the first data block with a first
` compression technique and second data block, second
` compression technique, so --
` Let's see. And then -- then we go down to
` claim -- what we label Claim -- or what I label Claim I,
` I should say: "Said compression storage occurs faster
` than said data stream is able to be stored."
` So by -- in my understanding, then said
` data stream includes a first and second block, and the
` said data stream thus is stored faster than could be
` stored in the received form; therefore, the first data
` block are stored faster.
` The other claim is more explicit. In
` Claim 1 in the other patent, that's more explicit;
` whereas, it specif- -- it says -- it does not use the
` word "data stream." It says specifically -- it says,
` "Wherein the first" -- what I'm labeling Claim F in my
`
`3 (Pages 6 to 9)
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`
`
`Page 10
`
`Page 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
`
`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
`
` deposition (sic) for the '908 patent, it says, "Wherein
` the first and second data stream blocks are stored on
` the memory device, and the compression and storage
` occurs faster than the first and said (sic) data blocks
` are able to be stored on the memory device in
` uncompressed form."
` So I would -- I would -- I would agree
` with that, though the language is a little bit
` different. I would agree with your statement.
` Q. (By MR. NOROOZI) Now, in answering my
` question, you were also looking at your declaration with
` respect to the '908 patent, right?
` A. That is correct.
` MR. NOROOZI: And so we'll mark that as an
` exhibit with the same exhibit number that it has in that
` proceeding.
` Q. (BY MR. NOROOZI) So I want to ask you a few
` more questions about the "faster than" limitation of the
` Claim 1s and how they work within the claim.
` And -- withdrawn.
` For purposes of both of the Claim 1s of
` the '530 and '908 patents, the "faster than" limitation
` requires the compression of both data blocks using two
` different techniques, plus the storage of those same two
` compressed data blocks occur faster than those two data
`Page 11
`
` blocks could be stored without any compression
` techniques or algorithms being applied to them, right?
` MR. SOMMER: Object to form.
` A. Could you restate your question? It was rather
` long.
` Q. (BY MR. NOROOZI) Sure. And the reason it's
` long is I'm trying to make sure that we're kind of
` capturing all the limitations that go with the "faster
` than" limitation in one place. Are you with me on that?
` A. I am, yes. Yes.
` Q. Okay. So for purposes of both Claim 1s of the
` '530 and '908 patents, the "faster than" limitation
` requires the compression of both data blocks using two
` different techniques, plus the storage of those same two
` compressed data blocks occur faster than those two data
` blocks would be stored without any compression
` techniques or algorithms being applied to them, right?
` MR. SOMMER: Object to form.
` A. Well, I'm not sure that that's -- that's
` completely true in exactly how the claims statement -- I
` would -- I would agree that Claim 1, each of these
` Claim 1s taken a whole -- taken as a whole apply --
` requires that two different data compression blocks are
` applied to Block Number 1 and Block Number 2.
` I would agree with the claim as a whole.
`
`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
`
`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
`
` That's very clear. And I would -- so I would also agree
` that because the limitation that -- let's see -- said
` compression and storage occur faster than said data
` stream is able to be stored on memory device in received
` form, that implies, along with the earlier statement in
` Claim 1, those two together imply that two different
` compression algorithms are used on two different blocks
` or can be used, I should say, on two different blocks
` and that the -- that the sum total of this process must
` allow for compression fast -- compression and storage
` that is faster than storage of uncompressed data alone.
` Q. (BY MR. NOROOZI) I just want to clarify if
` there was an aspect of what I articulated that you
` disagree with so that we make any of those issues clear
` on the record. So let me break it down, if I could,
` step by step. Withdrawn.
` Do you agree that the "faster than"
` limitation of Claim 1 of the '530 and Claim 1 of the
` '908 sets up a comparison with respect to two data
` blocks that compares the time it would take to store
` those two data blocks in uncompressed form versus the
` time it would take to store those two data blocks after
` compression and storage, right?
` MR. SOMMER: Object to form.
` A. I believe that -- that this limitation in the
`
`Page 13
`
` claim says that -- that it must be possible to compress
` these two data blocks and to store them in less time
` than it would take to store those same two data blocks
` in an uncompressed form.
` Q. (BY MR. NOROOZI) Okay. And when you say "it
` must be possible," you understand there's a difference
` between method claims and system claims, right?
` A. Yes.
` Q. And with respect to the method claims, the
` limitation must actually be met, right?
` A. Right. Sorry. I --
` MR. SOMMER: Object to form.
` Give me time to object.
` THE WITNESS: Okay. Sorry.
` A. Yes. It's -- this is -- this is a claims
` limitation that must be met. So what I should have said
` is I should have said to meet the requirements of the
` claim that the two blocks -- the compression and storage
` of those two blocks must result in a faster overall
` storage time than storing those two blocks uncompressed.
` So that is a limitation that must be met for the
` limitation of this claim to be fulfilled.
` And I -- I apologize if I -- if I
` misstated.
` Q. (BY MR. NOROOZI) No apology is needed. Thank
`
`4 (Pages 10 to 13)
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`
`
`Page 14
`
`Page 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
`
`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
`
` you, though, for clarifying that.
` Now -- withdrawn.
` For purposes of the "faster than"
` limitation of both of the Claim 1s of the '530 and
` '980 -- '908 patents, the resulting faster than storage
` must occur on the same storage device, right?
` MR. SOMMER: Object to form.
`(BY MR. NOROOZI) Why don't I rephrase that and
`Q.
` see if I can make it even clearer. Withdrawn.
` When we're talking about the "faster than"
` limitation of the Claim 1s of the '530 and '908 patents,
` and specifically talking about the storage aspect of the
` two data blocks in those Claim 1s, those two data blocks
` need to be stored on the same storage device, right? On
` one storage device?
` MR. SOMMER: Object to form.
`A. The claim language is "memory device," but that
` certainly could -- could be equated to a storage device.
`Q.
`(BY MR. NOROOZI) And with respect to the rest
` of my question, do you agree that for purposes of the
` "faster than" limitation, the two data blocks in
` question need to be stored on a single storage device or
` memory device?
`A.
`I agree that the claim says that a memory
` device is -- it says specifically "data accelerator is
`
`Page 15
`
` coupled" -- "is coupled to memory device," and it says
` in what we've -- what I've labeled Claim -- part C of
` Claim 1 of the '530 patent, and then it says that
` compressed stream is stored on said memory device. So,
` yes, I would agree that that -- that since the
` compressed stream is composed of two blocks, I would
` agree that those two blocks are stored on the same
` memory device.
`Q. And as I think you just said, the storage
` device on which the two data blocks are stored has to be
` the same one that would otherwise store the two
` uncompressed data blocks, right?
` MR. SOMMER: Object to form.
`A. According to Part I, said compression and
` storage occurs faster than said data stream is able to
` be stored on memory device in received form, so
` certainly with respect to Claim 1 of the '530 patent,
` that is -- that is correct.
` If I look at the other patent, on the
` memory device, the other patent -- what I've labeled
` Limitation G in Claim 1 of the other patent also says
` essentially the same thing. So to answer your question,
` yes.
`(BY MR. NOROOZI) The Claim 1s of the two
`Q.
` patents, the '530 patent and '908 patent, set up a
`
`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
`
`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
`
` comparison with respect to storing two uncompressed data
` blocks and two compressed data blocks, right?
` MR. SOMMER: Object to form.
`A. A comparison. Well, I mean, they certainly --
` the limitation clearly states that -- that -- clearly
` states two cases, and it gives a condition under which
` one case, storage -- compression and storage must be
` faster than the other case. So -- so I -- I think that
` that would -- by most people's definition, that would be
` a comparison, so, yes, I would agree with you.
`Q.
`(BY MR. NOROOZI) Now, for purposes of that
` comparison, the storage device or the memory device is a
` constant factor as between what happens with the two
` uncompressed data blocks and what happens with the two
` compressed data blocks, right?
` MR. SOMMER: Object to form.
`A. There is only one memory device referenced here
` and it is referenced consistently, so I would agree that
` that should be viewed as a constant factor.
`Q.
`(BY MR. NOROOZI) And so that means that the
` storage speed capabilities of the system at issue in the
` Claim 1s of the two patents is also a constant for
` purposes of the Claim 1s of the patents, right?
` MR. SOMMER: Object to form.
`A. Yes. So that -- so that the claim -- this
`
`Page 17
`
` limitation of Claim 1 -- I would agree with you that the
` limitation of Claim 1 then is relative to the specific
` memory device being evaluated. So you're -- you're
` contemplating a specific memory device, and then you are
` making that comparison.
`Q.
`(BY MR. NOROOZI) And similarly, the transfer
` speed capability from the data accelerator to the
` storage device is also constant for purposes of the two
` Claim 1s of the '530 and '908 patents, right?
` MR. SOMMER: Object to form.
`I would ask you to clarify that. What transfer
`A.
` speed are you referring to?
`Q.
`(BY MR. NOROOZI) I'm talking about the
` actual -- so -- withdrawn.
` When we talk about the storage speed
` capabilities, we're talking about some kind of a data
` write per amount of time rate, correct?
`A. For instance, a bandwidth, number of bits per
` second, is that what you're referring to?
`Q. Yes.
`A. Okay.
`Q. And so there's also -- withdrawn.
`And there's some limitation that all disk
` drives have as to how quickly they can write, for
` example, right?
`
`5 (Pages 14 to 17)
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`
`
`Page 18
`
`Page 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
`
`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
`
` A. Every device, not -- not all disk drives, but
` every memory storage device has some bandwidth
` limitation. So no matter what it is, there are some
` limitations to how fast it can accept data at its
` fastest speed.
` Q. And for purposes of -- withdrawn.
` There's also some physical hardware
` limitation in any system as to how quickly the data can
` move from where the data accelerator would be to the
` storage device itself, right?
` A. There's always some finite speed of the
` movement of data, given that one way or another the data
` is being moved by moving electrons in some sense. So
` the speed of light will govern some -- at some level.
` Whether or not that is at all significant depends --
` depends upon the exact details of the implementation.
` Q. And that is what I'm getting at. So what I
` mean is, depending on the hardware you've got going
` between where the data accelerator would be and where
` the storage or memory device would be, you could have a
` different maximum possible transfer rate for the data
` that comes from the accelerator to the memory device,
` true?
` A. So the hard -- I believe -- correct me if I am
` wrong, but I believe the hardware you're referring to is
`Page 19
`
` perhaps something like the hard disk controller, or --
` or something like that. I mean, I'm not sure -- when
` one is talking about generic storage, I think it's very
` difficult to generalize this. So I think you need to be
` a little bit more specific on what you're really
` referring to.
` Q. Okay. So let's assume an implementation of
` Claim -- of both of the Claim 1s of the '530 and '908
` patents where the data accelerator is not itself a part
` of the memory device or storage device. Are you with me
` so far?
` A. Okay.
` Q. And so there needs to be some transfer of data
` from the data accelerator to the memory or storage
` device in that scenario, right?
` A. Okay.
` MR. SOMMER: Object to form.
` A. Okay. Yeah. I can -- I can see how there can
` be some scenario where it was external to the -- to the
` hardware or to the hardware controller.
` Q. (BY MR. NOROOZI) And in that scenario, there
` would be some physical limitation in how quickly, in
` terms of data rate, the maximum speed that the data
` could move from the data accelerator to the memory
` storage device, right?
`
`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
`
`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
`
` MR. SOMMER: Same objection.
` A. It's possible there could be, depending upon --
` in this scenario, you know, you haven't told me anything
` about how these things -- you're saying that the
` accelerator is not part of, for example, the hard disk
` controller, where it would basically be effectively
` attached to the hardware. But you're not telling me
` what kind of -- of communications connection is --
` you're viewing as connecting that.
` Q. (BY MR. NOROOZI) And that's right. And so
` depending upon what the communications connection is
` between the accelerator and the storage device or the
` memory device, there could be a different maximum data
` transfer rate that is achievable, right?
` A. Well, certainly if your storage device were,
` you know, over in India and you're sending this through
` the Internet, then there's most definitely going to be a
` significant limitation on the -- you know, due to the
` communications channel. So I -- you know, there are
` many scenarios where you had a highly remote storage
` device where -- where you might have an issue.
` Q. And putting aside whether or not you would have
` any kind of an actual issue and assuming that the data
` accelerator is within the same physical piece of
` hardware as -- as -- as the storage device as well,
`Page 21
`
` meaning -- withdrawn.
` Assume, for example, we're talking about
` something like the laptop I have in front of me --
` A. Uh-huh.
` Q. -- right? You're with me so far?
` A. Yeah.
` Q. And in this laptop, we've got a storage device,
` right? A memory device?
` A. I hope so.
` Q. Yeah. It's safe to say, right?
` And -- withdrawn.
` And assume we also hypothetically have a
` data accelerator, something that meets the data
` accelerator claim definition within this laptop as well.
` Are you with me there?
` A. Okay.
` Q. And these two components are separated
` physically, they are not both within the memory device.
` Are you with me there?
` A. Okay.
` Q. There needs to be some connection between these
` two via hardware for data to move from the data
` accelerator to the memory device, right?
` A. Yes.
` Q. And depending on what the nature of that
`
`6 (Pages 18 to 21)
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`
`
`Page 22
`
`Page 24
`
`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
`
`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
`
` hardware connection is, there would be a physical
` limitation on how quickly the data that comes out of the
` data accelerator could move to the memory device, right?
`A. Did you say "there would be"? Did you say
` "there would be"? Did you use the word "would"?
`Q.
`I may have said "could," but I certainly meant
` "would." So let me rephrase it as "would."
`A. Okay. Well, I --
`MR. SOMMER: Object to form.
`I don't -- I don't agree with "would." I mean,
`A.
` there could be, depending upon your hardware
` implementation. Remember, you have to -- in this -- in
` the -- in the hypothetical scenario that you're
` describing, you have to look at what the relative
` storage speed of your storage media is versus the bus
` speed of your -- of your basically CPU bus, because
` that's more or less what you're using.
` But I would point out that -- that in one
` of the prior art references, Osterlund -- that is one
` reason why Osterlund talks about the wide multi- --
` multibit bus is to -- is to alleviate any concern about
` delays caused by such issues.
`Q.
`(BY MR. NOROOZI) Now, for purposes of the
` Claim 1s of the '530 and '908 patents, just as the
` storage device to which the data is being written is a
`Page 23
`
` constant, the hardware capabilities of the device to
` transfer data from the data accelerator to that storage
` device are also a constant, right?
` MR. SOMMER: Object to form.
`A. Well, I don't believe in Claim 1 of either of
` these two patents the issue of the speed of the
` transmission of the data, the compressed data, is -- is
` addressed. So I don't believe that there -- that
` Claim 1s, either of these Claim 1s, says anything about
` that.
`(BY MR. NOROOZI) And that's what I'm getting
`Q.
` at. So the Claim 1s set up a comparison between how
` quickly -- withdrawn.
` The Claim 1s have a "faster than"
` limitation, right?
` MR. SOMMER: Object to form.
`A. Yes. For instance, Claim 1 on the '530 patent
` says, "Said compression and storage occurs faster than
` said data stream is able to be stored on said memory
` device in said received form." So that is the
` limitation you're referring to, correct?
`Q.
`(BY MR. NOROOZI) Yes. And for purposes of
` that "faster than" limitation, the claims also talk
` about a data accelerator, right?
` MR. SOMMER: Object to form.
`
`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
`
`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
`
`A. The claims introduce the -- the description of
` a device that is termed a "data accelerator."
`Q.
`(BY MR. NOROOZI) And it is that data
` accelerator that is supposed to make it such that
` compression and storage of the two compressed data
` blocks occurs faster than uncompressed storage of those
` same two data blocks would occur, right?
`A. Well, you know, as I'm sure you've read in my
` declaration, you will know that -- that it's not clear
` that just the term "data accelerator" alone, that the
` data accelerator has to actually accelerate the storage.
` But when combined with I, when combined with
` Limitation I, then, combined with Limitation I, the data
` accelerator is required to store faster, store the two
` blocks in compressed form, compressed form plus storage
` time should be faster than storing them in uncompressed
` form alone.
`Q. So to give an example to clarify what I'm
` getting at, let's say that you had two identical
` versions of my laptop side by side, right?
`A. Okay.
`Q. And you gave both of those laptops the same
` identical two input data blocks. Are you with me so
` far?
`A. Okay.
`
`Page 25
`
`Q. And one of them had a data accelerator and the
` other one did not, correct? I mean, are you with me so
` far there?
`A.
`I guess, yes.
`Q. And so in that scenario, if we allowed the
` process to run and we -- and we clocked the time until
` those two data blocks in each of the two laptops were
` finished storing onto the memory device, we would have a
` comparison between the time it takes for the two data
` blocks to be stored in uncompressed form versus
` compression and storage, right?
` MR. SOMMER: Object to form.
`A. Hypothetically, assuming that there were no
` delays getting your data blocks -- getting your
` arbitrary data blocks into your laptop and -- or any
` other -- other untoward delays, if everything else was
` equal and we just had some system where we had two
` blocks of data, two -- two systems that were identical
` in every manner except for one had the data accelerator
` and one did not, we took two identical blocks, put them
` through one with, put them through one without, then you
` should see some sort of a difference.
`Q.
`(BY MR. NOROOZI) And if you do, then you would
` have a data accelerator under the Claim 1s, right?
` MR. SOMMER: Object to form.
`
`7 (Pages 22 to 25)
`
`TransPerfect Legal Solutions
`212-400-8845 - Depo@TransPerfect.com
`
`
`
`Page 26
`
`Page 28
`
`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
`
`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
`
` A. Well, again, you spent all this time pointing
` out that -- that -- you know, and you're discussing the
` main systems bus -- you're pointing out that there could
` potentially be other sources of latency in the
` transmission that are not accounted by -- by Limitation
` I of Claim 1. So I'm not sure this hypothetical
` really -- really matches the claim precisely.
` Q. (BY MR. NOROOZI) So we agreed earlier that the
` capabilities -- the write capabilities of the memory
` device in question in the Claim 1s would remain a
` constant, right?
` A. Yes. The number of bits per second, either on
` the average or at -- on the -- at the peak for a given
` memory device, is going to remain a constant.
` Q. And for purposes of the comparison of
` compressed form versus uncompressed form and the -- and
` the -- and the time it takes to get to storage,
` finishing storage in the memory device, do you also
` agree that the Claim 1s maintain the transfer capability
` of the -- of the system in question the same? They
` don't change the transfer capability as between
` compressed form and uncompressed form?
` MR. SOMMER: Object to form.
` A. What do you mean by the "transfer capability"?
` Q. (BY MR. NOROOZI) For example, when we're
`
`Page 27
`
` making the comparison that is set up by Claim 1 of the
` '530 and Claim 1 of the '908 and we've got a comparison
` between storing uncompressed data and compressed data,
` we're not changing the physical buses that run from the
` accelerator to the storage device for purposes of
` determining whether compression and storage are faster
` than uncompressed storage, right?
` MR. SOMMER: Object to form.
` A. I mean, I don't think that -- I don't think
` that's relevant to this claim because the claim doesn't
` talk about the transmission process.
` Q. (BY MR. NOROOZI) And so the claim does talk
` about compression, right?
` A. Of course, yes.
` Q. And the buses have nothing to do with the
` actual compression, right?
` A. I don't think you can make a blanket statement
` like that. I mean, if your compression algorithm were
` implemented on a parallel processor with a parallel
` processor connected by buses, then the buses would
` influence the compression process itself.
` Q. The -- the compression speed is not affected by
` the buses, right?
` A. Again, if you had a parallel implementation
` where you used the buses to -- or some bus to shunt data
`
`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
`
`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
`
` between the processors, then that bus speed would affect
` the speed of the compression algorithm.
` Q. You're talking about having two processors?
` A. Or more. You could have a -- many -- you know,
` many, for example, whatever.
` Q. Okay.
` A. But my point is that I can't accept a blanket
` statement like that without more restrictions.
` Q. And in the comment you were making, were you
` assuming that there would be one compressor within the
` system you were talking about or more than one
` compressor?
` A. Which comment?
` Q. When you were talking about if you've got buses
` between processors.
` A. No. If you had -- you can implement a single
` compression algorithm using multiple processors. Okay?
` So -- so -- and in the case, of course, as -- as pointed
` out by Franaszek where he uses multiple compression
` algorithms, one can use multiple parallel processors to
` implement multiple algorithms to more efficiently do
` your -- do the test required to choose the best
` algorithm.
` So my point is that to shunt data around
` to those parallel