`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`10
`
`11
`
`12
`
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`25
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
` BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`Page 158
`
`ORACLE CORPORATION, §
`NETAPP INC. AND HUAWEI §
`TECHNOLOGIES CO., LTD. §
` §
` Petitioners, § IPR2014-01197
` § IPR2014-01207
`VS. § IPR2014-01209
` §
` §
`CROSSROADS SYSTEMS, §
`INC. §
` §
` Patent Owner. §
`
` ORAL AND VIDEOTAPED DEPOSITION OF
` JOHN LEVY, PH.D.
` JULY 16, 2015
` CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` VOLUME 2 OF 2
`
` ORAL AND VIDEOTAPED DEPOSITION OF JOHN LEVY, PH.D.,
`produced as a witness at the instance of the Petitioners
`and duly sworn, was taken in the above styled and numbered
`cause on Thursday, July 16th, 2015 from 9:01 a.m. to
`2:04 p.m., before Tamara Chapman, CSR, RPR, CCR (LA) in
`and for the State of Texas, reported by computerized
`stenotype machine, at the offices of Sprinkle IP, 1301
`West 25th Street, Suite 408, Austin, Texas.
`
`Job No: 95252
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 1
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 159
`
` A P P E A R A N C E S
`
`FOR THE PETITIONERS ORACLE CORPORATION, NETAPP INC. AND
`HUAWEI TECHNOLOGIES CO., LTD.:
` Mr. Greg Gardella, Esq.
` OBLON MCCLELLAND MAIER & NEUSTADT
` 1940 Duke Street
` Alexandria, Virginia 22314
`
` -
` Mr. Aaron Huang, Esq.
` WEIL GOTSHAL & MANGES
` 201 Redwood Shores Parkway
` Redwood Shores, California 94065
`
`FOR THE PATENT OWNER CROSSROADS SYSTEMS, INC.:
` Mr. James Hall, Esq.
` BLANK ROME
` 700 Louisiana
` Houston, Texas 77002
`
`ALSO PRESENT:
` Ms. Rachel MacGuire - Oracle Senior Corporate Counsel
` Mr. Brent Kirby - The Videographer
`
`TSG Reporting - Worldwide 877-702-9580
`
`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
`
`Oracle Ex. 1219, pg. 2
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
` * * *
` EXAMINATION INDEX
`
`Page 160
`
` Page
` BY MR. GARDELLA............................ 161
`
` BY MR. HALL................................ 186
`
` BY MR. GARDELLA............................ 215
`
` * * *
`
` INDEX OF EXHIBITS
`
` Page
` Exhibit 1217............................... 186
` Telephone conference with The Panel
` 7/16/15
`
` * * *
`
` INDEX OF PREVIOUS EXHIBITS
`
` Exhibit 1018............................... 208
` Exhibit 1213............................... 188
` Exhibit 2053............................... 186
`
`TSG Reporting - Worldwide 877-702-9580
`
`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
`
`Oracle Ex. 1219, pg. 3
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 161
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` THE VIDEOGRAPHER: This is Tape No. 1.
`We're recording and on the record. 9:01 a.m.
` EXAMINATION
`BY MR. GARDELLA:
` Q. Welcome back, Dr. Levy.
` A. Thank you.
` Q. Did you speak with counsel about the substance of
`your testimony since we closed yesterday?
` A. No, I did not.
` Q. I'd like to next ask some questions about the
`Hirai reference. Do you have that in front of you?
` A. I will in a moment.
` Okay.
` Q. So -- and my questions are going to relate to the
`create and delete commands and the applicability that they
`might have to certain activities.
` A. All right.
` Q. So in the context of partition control in the
`Hirai system, is it fair to say that an administrator
`station on the host side could use create and delete
`commands?
` MR. HALL: Objection; foundation, form.
` Q. (BY MR. GARDELLA) Do you understand the
`question?
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 4
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 162
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` A. Well, let's see.
` Q. Well, do you understand the question, though,
`before we proceed?
` A. I am going to address that.
` Q. Okay. Sorry.
` A. Hirai does not actually talk about create and
`delete commands. He -- he only dis- -- says a personal
`computer can read, write, create and execute with
`partition 1 and so on. If that action of a personal
`computer, read, write, create and execute, is what you're
`referring to, then we can proceed from there.
` Q. Just to make sure we're on the same page, what
`paragraph of Hirai are you reading from?
` A. That's in Paragraph 13.
` Q. So how do you understand that sentence in -- in
`terms of what read, write, create and execute are
`referring to?
` MR. HALL: Objection; form.
` A. Well, since the access rights spoken about in
`Paragraph 12, which include, read, write, create, delete
`and execute, are well-known file system access rights, I
`understand the statement that a personal computer 2 can
`read, write, create and execute with Partition 1 to mean
`that a program running in personal computer -- I said 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 5
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 163
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`can perform those types of file access operations in that
`partition.
` Q. (BY MR. GARDELLA) In Hirai, one of the personal
`computers can serve as an administrative station or
`terminal. Correct?
` MR. HALL: Objection; foundation.
` A. Perhaps you can point me to a section of Hirai
`that talks about that.
` Q. (BY MR. GARDELLA) I don't believe there is any
`express discussion of it. My question is, rather, you
`know, given the overall system design and architecture
`described in Hirai, is it reasonable to infer that
`administrative functions could be carried out from one of
`the personal computers?
` MR. HALL: Objection; form.
` A. Well, not necessarily. I mean, there's really no
`discussion of how the partitions are defined or managed.
` Q. (BY MR. GARDELLA) And -- and I guess that's my
`point, Dr. Levy. Isn't it fair to say that in the context
`of a system such as that described in Hirai, an
`administrator performing administrative duties at one of
`the personal computers could utilize, create and delete
`commands to create and delete partitions on the storage
`devices?
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 6
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 164
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` MR. HALL: Objection; form, foundation.
` A. Well, there are -- you know, there are no create
`and delete commands discussed in Hirai. But in spite of
`that, I think also that -- if reading Hirai we come to the
`point of view that the magnetic disk sharing device is, in
`fact, a -- is something related to a file server, one can
`imagine that that server itself could have an
`administrator that does this kind of administration
`without necessarily using the personal computers attached.
` Q. (BY MR. GARDELLA) Okay. So do I understand you
`to be saying that the administrative functions could be
`carried out either at sharing device 3 or at one of the
`personal computers, 1 or 2?
` MR. HALL: Objection; foundation.
` A. Well, Hirai makes no suggestion of either one of
`those. I was just saying that it could easily be done by
`a file server itself with a user interface or somewhere
`else so that -- since Hi doesn't -- Hirai doesn't address
`that, there's really no reason to understand it as being
`one or the other.
` Q. (BY MR. GARDELLA) Okay. So, again, my question,
`which I -- I don't think you're answering quite yet, is
`whether the administrative functions that we're talking
`about could be performed at the personal computers. 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 7
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 165
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`let me reformulate the question.
` Can you identify any reason why, in the context
`of Hirai, that such administrative functions could not be
`performed by an administrator through personal computers 1
`or 2?
` MR. HALL: Objection; form, foundation.
` A. Well, one potential reason would be that since
`the purpose of the partition control table and its
`functions is to prevent illegal access from the personal
`computers, it may be necessary to set up that partition
`control table before any personal computer attached could
`have any kind of access. That would weigh strongly in
`favor of having an administrator directly connected to the
`sharing device.
` Q. (BY MR. GARDELLA) Depending on the policies that
`the administrator or a user of the Hirai system wanted to
`enforce, would you agree that it would be possible to have
`an administrator utilize create and delete commands at a
`PC station to create and delete partitions on the storage
`devices?
` MR. HALL: Objection; form, foundation.
` A. Well, there are no create and delete commands.
`And, furthermore, we're way beyond anything that Hirai
`discusses in terms of an administrator, so --
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 8
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 166
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` Q. (BY MR. GARDELLA) Right.
` A. -- it's pure speculation.
` Q. Yeah. I understand that it's not expressly
`described in Hirai, which is one of the reasons why I'm
`asking you the question. We're, you know, trying to
`establish through this discussion what types of things
`would be reasonable or not reasonable in the context of
`the Hirai system. And I'd like you to assume, for
`purposes of my next question, that in the Hirai system
`there are control in- -- excuse me -- there are create and
`delete commands which correspond to the create and delete
`permissions. Would you please assume that?
` MR. HALL: Objection; foundation.
` A. I'm willing to make that literal assumption, but
`you'll have to tell me what they do.
` Q. (BY MR. GARDELLA) Well, my question kind of gets
`to that premise. If a system such as that described in
`Hirai has create and delan- -- delete commands which
`correspond to the create and delete permissions, an
`administrator at one of the PCs could use such commands to
`create or delete partitions on the storage devices. Is
`that fair?
` MR. HALL: Objection; form, foundation.
` A. Well, I don't know of a create partition command
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 9
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 167
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`or a delete partition command that would be -- that exists
`or would be relevant.
` Q. (BY MR. GARDELLA) I'd ask you to assume that
`those commands exist. If those commands exist, could such
`commands be used by an administrator at one of the PCs in
`a system such as that described --
` MR. HALL: Objection --
` Q. (BY MR. GARDELLA) -- at Hirai --
` MR. HALL: Objection; form.
` Q. (BY MR. GARDELLA) -- or in Hirai? Excuse me.
` MR. HALL: Objection; form, foundation.
` A. So you're asking me to assume that there is a
`create partition command?
` Q. (BY MR. GARDELLA) Yes.
` A. And that there exists a delete partition command?
` Q. (Nods.)
` A. And perhaps you can enlighten me as to what
`they're supposed to do.
` Q. You can assume they do what their names imply,
`that they create and delete partitions. Such commands
`could be used by an administrator at one of the PCs in the
`context of a system such as that described in Hirai.
`Correct?
` MR. HALL: Objection; form, foundation.
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 10
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 168
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` A. Well, you'd -- before such a command would be
`relevant, you'd have to have also assumed that an
`administrator at the PC, first of all, had access to the
`disk-sharing system, that it also had done the aggregation
`of the disks required before the partitioning is going to
`be done, which is what Hirai describes, and possibly other
`items, such as setting up file systems. There's a whole
`bunch of background required before "partition" could be
`defined.
` Q. Those assumptions all sound reasonable to me, so
`please make those assumptions.
` A. Okay. Now what?
` Q. So then what's the answer to the question?
` Would you like it back?
` A. I'd like -- yes.
` Q. Making those assumptions, create and delete
`commands could be used by an administrator at one of the
`PCs in the context of a system such as that described in
`Hirai. Is that correct?
` MR. HALL: Objection; form, foundation.
` A. So let me just enumerate the assumptions I'm
`making here first.
` We're assuming that this magnetic disk sharing
`device is running and that PCs have access to it before
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 11
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 169
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`the access rights to partitions are set up. Then an
`administrator is operating on one of these PCs and has
`access to it and the magnetic disk sharing device, that
`the configuration of the physical disk attached to the
`sharing device is visible, and that the administrator has
`arranged the aggregation of those disks into a larger,
`logical or virtual volume.
` And now these putative create partition commands
`and delete partition commands exist and would do something
`analogous to what a volume manager would do in a normal
`file server. Under those assumptions, it's possible that
`this could be managed from a PC attached to that server.
` Q. (BY MR. GARDELLA) Let's move on now to the
`execute command. In the context of Hirai, how would the
`sharing device 3 enforce an execute command?
` A. Well, you're going to have to tell me what an
`execute command does.
` Q. Well, let's start there. What is your
`understanding of what the execute access right described
`in Paragraph 12 of Hirai permits?
` A. The execute access right in a file system does --
`says that the user who was accessing that file is allowed
`to have it read into the system of the user and then
`executed by the processor of that system.
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 12
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 170
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` Q. Okay. And that, again, is -- well, strike that.
` In the Hirai system, the sharing device 3 has no
`ability to monitor or control the execution activities of
`PCs 1 or 2, does it?
` A. Well, I think if you look at the handling of the
`execute access right in the context of the NFS system, for
`example, that access right would be -- could be checked by
`both the file server and the system that -- remote system
`that has requested the access to that file.
` So while the execution of the file would only
`take place in the local system that requested the file,
`the access right may well be checked by the file server.
` Q. Of what utility would that be, given that PCs 1
`or 2 are going to be executing the file locally?
` A. It would prevent -- it would remove the need to
`transmit a file for which the user doesn't have permission
`to execute.
` Q. In this context, I mean, files are simply blocks
`of data. Is that fair?
` MR. HALL: Objection; form.
` A. At base, they are simply blocks of data. But in
`the case of a file which is proposed to be executed, it
`has to be a code file, a file that can be executed.
` Q. (BY MR. GARDELLA) After a file under 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 13
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 171
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`interpretation of Hirai is sent to PC 1 or 2, how would
`the sharing device 3 prevent PC 1 or 2 from executing that
`file in the event they did not have execute permission?
` A. Well, if -- if the Hirai system functions the way
`the NFS file server works, both the local system that's
`requesting the file and the file server, which in this
`case would be the magnetic disk share picking (phonetic)
`device, would check the user's access rights to determine
`whether execute permission is allowed.
` So it would have been checked before it sent.
`But if for some reason you were trying to fool it -- you
`could send in a read request without saying that it's
`going to be executed, I suppose, but that wouldn't be the
`way -- that wouldn't be kosher in this context of NFS.
`The operating systems were always honest with each other.
` Q. You would agree, then, that in the event PC 1 or
`2 submitted a read request for data which was executable
`and had read permissions at the sharing device 3 and
`thereafter received the requested data, that in that
`circumstance the sharing device 3 would have no ability to
`prevent PC 1 or 2 from executing the data irrespective of
`whether PC 1 or 2 had execute permission?
` MR. HALL: Objection; form, foundation.
` A. It is theoretically possible for a -- an
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 14
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 172
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`operating system to overcome the access permissions of an
`executable file by having it masquerade as a readable file
`but not an execute file. So what you suggest is -- is
`conceivable.
` Q. (BY MR. GARDELLA) And would you agree that
`there's nothing in Hirai which would prevent that
`possibility?
` MR. HALL: Objection; form, foundation.
` Q. (BY MR. GARDELLA) Would you agree that there's
`nothing expressly taught in Hirai which would preclude
`that?
` A. Well, there's -- there's precious little
`expressly taught in Hirai about what the PCs are doing, so
`that is not expressly taught.
` Q. So you agree with me that there is nothing
`expressly taught in Hirai which would preclude what I
`proposed?
` MR. HALL: Objection; form.
` A. There's nothing expressly taught in Hirai that
`would preclude a user on a PC execu- -- executing a read
`request on a file that would be sent back and subsequently
`executed.
` Q. (BY MR. GARDELLA) Even though the PC did not
`have execute permissions for that 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 15
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 173
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` A. Even though the user did not have -- have execute
`permission on that file.
` Q. Let's turn to Bergsten next. Do you have
`Bergsten --
` A. Yes.
` Q. -- before you?
` Let's go to Claim 38 first. Could you read
`Claim 38 into the record, please.
` A. Okay. So Claim 38 is dependent claim --
`depending from Claim 37, which depends from Claim 36,
`which depends from Claim 35, just to note that.
` Q. And you can feel free to review those to the
`extent you think is necessary. I don't know that it will
`be necessary in the context of my question.
` A. Okay.
` Q. But could we start by reading Claim 38 into the
`record?
` A. Yes. Claim 38, "A system according to Claim 37,
`each of the storage devices further comprising translation
`means for representing the storage arrays as at least one
`virtual device to the host processing systems."
` Q. That language contemplates and suggests that
`there may be more than one virtual device. Is that
`correct?
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 16
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 174
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` MR. HALL: Objection; calls for a legal
`conclusion.
` A. It -- I -- I think that it leaves open that
`possibility.
` Q. (BY MR. GARDELLA) Let's turn next to Column 8,
`Line 26.
` A. If I could just elaborate on the last answer.
` Q. Absolutely.
` A. It appears to me that the -- while the Claim 38
`leaves open the possibility that there's more than one
`virtual device, it doesn't seem to suggest that there's a
`different represe- -- representation to any individual
`host, that whatever representation there is is to the host
`processing systems.
` Q. Are you ready to turn to Column 8 now?
` A. Yes.
` Q. Do you see the sentence in Line 26 of Column 8
`which begins with the phrase "Figure 6 illustrates"?
` A. Yes.
` Q. Could you please read that and the subsequent two
`sentences into the record, please.
` A. "Figure 6 illustrates a routine for performing
`these standard checks."
` Do you want the next one as well?
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 17
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 175
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` Q. The next two, please.
` A. Sorry.
` "It is first determined, in step 601, whether the
`storage device identified by the ho-" -- "by the host
`exists. As noted above, this device will be a virtual
`device."
` Q. Given that this passage describes, in the first
`instance, checking determine -- strike that.
` Given that this passage describes as the first
`step of the process checking to determine whether the
`identified storage device exists and that that storage
`device is a virtual device, doesn't this also suggest and
`contemplate the existence of more than one virtual device?
` MR. HALL: Objection; foundation.
` A. It does suggest that the storage device
`identified by a host -- it's conceivable that there may be
`more than one storage device that could be identified by a
`host and that that such would be a virtual device.
` Q. (BY MR. GARDELLA) Would you agree that if a
`system such as that taught in Bergsten was hardwired or
`configured with a single virtual device, there would be no
`advantage or utility in performing this check as the first
`step of the process described in Figure 6?
` MR. HALL: Objection; form, foundation.
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 18
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 176
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
` A. Well, since this section here is discussing
`standard checks that are done by a storage controller, I
`think that it's really just pointing out that any
`controller would have to verify that a access request is
`to something that exists.
` Q. (BY MR. GARDELLA) And that check would be
`superfluous in the context of a system which was hardwired
`to have only one virtual device. Would you agree with
`that?
` A. No.
` MR. HALL: Objection; form, foundation.
` A. No, since the form in which requests come in have
`to designate some kind of device. Even if there's only
`one, one would want to check to make sure that was what
`was being addressed.
` Q. (BY MR. GARDELLA) So is it fair to say, then,
`that this passage of Bergsten permits the use of multiple
`virtual devices?
` MR. HALL: Objection; foundation.
` A. I would say that this passage of Bergsten does
`not eliminate that possibility, although it doesn't
`particularly suggest it.
` Q. (BY MR. GARDELLA) Can you point to anything in
`Bergsten, sitting here today, which precludes the use of
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 19
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 177
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`multiple virtual devices?
` MR. HALL: Objection; scope, form,
`foundation.
` A. (Reviewing document.)
` Well, I think since Bergsten describes virtual
`address as being what comes in and gives great detail
`about how to map a virtual address into logical and then
`physical device addressing, a person of ordinary skill in
`the art reading this would understand that the virtual
`addresses are in a unitary or singular -- single address
`space.
` I guess I would expect that if Bergsten was
`anticipating multiple virtual addresses, he would have
`suggested that that virtual address space could be one of
`several, but it doesn't seem to be described that way.
` Q. (BY MR. GARDELLA) I understand that you're
`inferring from the overall description of Bergsten that
`Bergsten intended, as the preferred implementation, that
`there be a single virtual device. However, that is not my
`question.
` My question is whether there is any teaching in
`Bergsten which would preclude the implementation of the
`Bergsten system with multiple virtual devices.
` MR. HALL: Objection; scope, 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 20
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 178
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`foundation.
` A. Well, while -- I don't think I could find
`language that precludes that possibility. But since
`Bergsten equates the host address with a virtual address
`and then shows how to map that virtual address, it leads
`to -- leads me, anyway, to the conclusion that there's
`only one virtual address, and it's the address provided by
`the host.
` Q. (BY MR. GARDELLA) But there could be multiple
`host addresses and multiple virtual devices. Correct?
` MR. HALL: Objection -- objection;
`foundation.
` A. Well, I hadn't considered that before, but I
`don't see anything at the moment that would necessarily
`preclude it.
` Q. (BY MR. GARDELLA) So I'd like to shift gears a
`bit and discuss the combination of Bergsten and Hirai.
` Now, I'm going to ask you to accept, for purposes
`of this next question, that the combination does provide
`multiple LUNs of virtual devices. I'd like to understand
`that under that assumption -- or, rather, I'd like to
`understand whether under that assumption there are
`additional issues with the combination.
` So the question is this. If you assume that the
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 21
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 179
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`combination of Bergsten and Hirai provides multiple LUNs
`or virtual devices, then is it your view the only
`remaining problem with the combination is that Bergsten
`doesn't identify the host?
` MR. HALL: Objection; form, foundation.
` A. No.
` Q. (BY MR. GARDELLA) Please identify any additional
`problems that would exist with the combination.
` MR. HALL: Objection; form, foundation.
` A. Well, first of all, of course, you're asking me
`to assume there's something about LUNs which is not
`disclosed here. But let me leave that for the moment and
`try to answer your question.
` When -- if one were to combine Hirai as modified,
`according to your assumption --
` Q. (BY MR. GARDELLA) Uh-huh.
` A. -- with Bergsten in the way in which it was
`suggested by Dr. Chase, which would be at the logical
`level, you would still have the problem that a host could
`not detect what access controls were being applied to its
`storage at the logical level because the logical level is
`not visible to the host.
` Q. And why is it essential that the host have
`visibility into the access controls which are being
`
`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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 22
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 180
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`applied to it?
` MR. HALL: Objection; form.
` A. Because, as I've pointed out in -- somewhere --
`let's see here.
` (Reviewing document.)
` Q. (BY MR. GARDELLA) May I ask which paragraph of
`your declaration you're making reference to?
` A. I will -- will tell you when I find them, yes.
` Well, I -- in -- in Paragraphs 132 through 135 --
`actually, 136 -- explain a complication by the combination
`where Hirai's file access permissions, that those same
`problems would occur. And -- and the reason for those
`problems is that the host can't determine which portions
`of storage are -- access is permitted or denied because
`it's being applied to the logical disk level. And so
`those same problems would come up under the assumptions
`you've given me.
` Q. Well, would you agree that a combined system
`which was subject to the shortcomings set forth in
`Paragraphs 132 to 136 of your declaration would still be
`useful in certain applications and for certain purposes?
` MR. HALL: Objection; form, foundation.
` A. No.
` Q. (BY MR. GARDELLA) So it's your view that 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
`
`TSG Reporting - Worldwide 877-702-9580
`
`Oracle Ex. 1219, pg. 23
`Oracle, et al. vs. Crossroads
`IPR2014-01207
`(CONFIDENTIAL)
`
`
`
`Page 181
` J. LEVY-7/16/15-CONFIDENTIAL PROTECTIVE ORDER MATERIAL
`system in which the host did not have visibility into the
`access control permissions which were being applied to it
`would have no utility?
` MR. HALL: Objection; form.
` A. As I've explained in Paragraph 136 where even --
`even assuming you could have per-host access rights,
`there's no reasonable expectation of success of such a
`system being functional or useful.
` Q. (BY MR. GARDELLA) Okay. Please explain why that
`is.
` A. Well, it -- let's just say it's very awkward when
`you have a system which is accessing storage to have its
`commands fail for no reason that the host can determine.
` Q. Commands fail in other computing contexts without
`causing failure or inoperability of the system. Correct?
` MR. HALL: Objection; form, foundation.
` A. While there's a great deal of effort spent in
`systems dealing with failures which are unavoidable, this
`is an entirely avoidable type of failure that no -- I
`believe no reasonable person of skill in the art would