ACTIFIO, INC. v. DELPHIX CORP.

Filing 117

Judge Denise J. Casper: ORDER entered. MEMORANDUM AND ORDER. (Maynard, Timothy)

UNITED STATES DISTRICT COURT
DISTRICT OF MASSACHUSETTS
_________________________________________
)
)
)
)
)
)
)
)
)
)
)

ACTIFIO, INC.,
Plaintiff,
v.
DELPHIX CORP.,
Defendants.
________________________________________

Civil Action No. 14-13247-DJC

MEMORANDUM AND ORDER
CASPER, J.
I.

December 11, 2015

Introduction
Plaintiff Actifio, Inc. (“Actifio”) alleges that Defendant Delphix Corp. (“Delphix”)

infringes certain claims of U.S. Patent Nos. 8,299,944 (“’9944 patent”), 8,788,769 (“’769
patent”) and 8,904,126 (“’126 patent”) (collectively, the “Patents-in-Suit”). The parties now
seek construction of five disputed claim terms. After extensive briefing and a Markman hearing,
the Court’s claim construction follows.
II.

Patents-in-Suit
This lawsuit involves patents addressing virtual data management, storage and backup.

D. 67-1 (’9944 patent), 67-2 (’769 patent), 67-3 (’126 patent); see Actifio, Inc. v. Delphix Corp.,
No. 14-cv-13247-DJC, 2015 WL 1243164, at *4 (D. Mass. Mar. 17, 2015). Actifio alleges that
Delphix’s “Agile Data Platform” products and services infringe one or more of the claims of the
Patents-in-Suit.

D. 67 ¶¶ 18, 22-23, 35-36, 41-42.

The Patents-in-Suit share similar

specifications and incorporate each other by reference. ’9944 patent, 1:16-23; ’769 patent, 1:1518, 1:30-33; ’126 patent, 1:16-19, 36-38. All the Patents-in-Suit were filed on November 16,
1

2010. D. 67-1 at 1; D. 67-2 at 1; D. 67-3 at 1. The ’9944 patent issued on October 30, 2012, D.
67 ¶ 15, the ’769 patent issued on July 22, 2014, id. ¶ 16, and the ’126 patent issued on
December 2, 2014, id. ¶ 17.
III.

Procedural History
Actifio instituted this action on August 6, 2014. D. 1. On August 22, 2014, Delphix

moved to dismiss, stay or transfer the action to the Northern District of California. D. 7. The
Court denied that motion, D. 54, and a history of the litigation between the parties is outlined in
the Court’s Order on that motion. See Actifio, 2015 WL 1243164, at **2-3. Actifio filed an
assented-to amended complaint on April 14, 2015. D. 67. After claim construction briefing, the
Court held a Markman hearing on November 19, 2015 and took the matter under advisement. D.
110.

Later that day, without leave of the Court, Delphix filed a “Supplemental Claim

Construction Submission” which included the arguments it presented at the Markman hearing.
D. 111. Actifio filed a reply and requested Delphix’s submission be stricken, D. 112, and
Delphix opposed, D. 115. Considering the parties have had an opportunity to respond to each
other’s post-hearing filings, the Court considers all of the submissions here and denies Actifio’s
motion to strike the supplemental submission filed by Delphix.
IV.

Standard of Review
As established by the Supreme Court, construction of disputed claim terms is a question

of law for the Court. Markman v. Westview Instruments, Inc., 517 U.S. 370, 372 (1996). This
Court must construe “the meaning that the term would have to a person of ordinary skill in the
art in question at the time of . . . the effective filing date of the patent application.” Phillips v.
AWH Corp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) (en banc). In doing so, the Court must look
to “the words of the claims themselves, the remainder of the specification, the prosecution

2

history, and extrinsic evidence concerning relevant scientific principles, the meaning of technical
terms, and the state of the art.” Id. at 1314 (quoting Innova/Pure Water, Inc. v. Safari Water
Filtration Sys., Inc., 381 F.3d 1111, 1116 (Fed. Cir. 2004)) (internal quotation marks omitted).
A.

The Claims

Claim construction begins with the words of the claims themselves where “the claims of
a patent define the invention to which the patentee is entitled the right to exclude.” Id. at 1312
(quoting Innova, 381 F.3d at 1115) (internal quotation marks omitted). Claims “are generally
given their ordinary and customary meaning” and can “provide substantial guidance as to the
meaning of particular claim terms.” Id. at 1312, 1314 (quoting Vitronics Corp. v. Conceptronic,
Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996)) (internal quotation marks omitted). “[T]he context in
which a term is used in the asserted claim can be highly instructive.” Id. at 1314. A claim itself
may provide the means for construing a term where, for instance, the claim term is consistently
used throughout the patent. Id. As such, “the meaning of a term in one claim is likely the
meaning of that same term in another.” Abbott GmbH & Co., KG v. Centocor Ortho Biotech,
Inc., No. 09-cv-11340-FDS, 2011 WL 948403, at *3 (D. Mass. Mar. 15, 2011) (citing Phillips,
415 F.3d at 1314). Notably, “the presence of a dependent claim that adds a particular limitation
gives rise to a presumption that the limitation in question is not present in the independent
claim.” Phillips, 415 F.3d at 1315.
B.

The Specification

Claims are not to be read alone, but “are part of ‘a fully integrated written instrument,’
consisting principally of a specification that concludes with the claims.” Id. (citation omitted).
“Usually, [the specification] is dispositive; it is the single best guide to the meaning of a disputed
term.”

Id. (quoting Vitronics, 90 F.3d at 1582) (internal quotation mark omitted).

3

The

specification, as the patentee’s description of the invention, defines “the scope and outer
boundary” of the claims and, thus, “claims cannot be of broader scope than the invention that is
set forth in the specification.” On Demand Mach. Corp. v. Ingram Indus., Inc., 442 F.3d 1331,
1338-40 (Fed. Cir. 2006). In looking to the specification in interpreting the meaning of a claim,
the Court must be careful not to “import[ ] limitations from the specification into the claim.”
Phillips, 415 F.3d at 1323. This standard may “be a difficult one to apply in practice,” id., but
“[t]he construction that stays true to the claim language and most naturally aligns with the
patent’s description of the invention will be, in the end, the correct construction,” id. at 1316
(quoting Renishaw PLC v. Marposs Societa’ per Azioni, 158 F.3d 1243, 1250 (Fed. Cir. 1998)).
C.

The Prosecution History

After looking to the claims themselves and the specification, “a court should also
consider the patent’s prosecution history, if it is in evidence.” Id. at 1317 (citation and internal
quotation marks omitted). The prosecution history, as evidence of how the inventor understood
the patent, “can often inform the meaning of the claim language by demonstrating how the
inventor understood the invention and whether the inventor limited the invention in the course of
prosecution, making the claim scope narrower than it would otherwise be.” Id. (citing Vitronics,
90 F.3d at 1582-83). Because the prosecution history “often lacks [] clarity” it is “less useful for
claim construction purposes” and is given less weight than the claims and the specification. Id.
D.

Extrinsic Evidence

Finally, the Court may also consider extrinsic sources which can aid in educating the
Court “regarding the field of the invention” and in determining “what a person of ordinary skill
in the art would understand claim terms to mean.” Id. at 1319. Dictionaries and treatises often
assist courts in understanding the underlying technology and “in determining the meaning of

4

particular terminology to those of skill in the art of the invention.” Id. at 1318. In general,
extrinsic evidence is considered “less reliable than the patent and its prosecution history in
determining how to read claim terms,” id., and thus “is less significant than the intrinsic record in
determining the legally operative meaning of claim language,” id. at 1317 (citations and internal
quotation marks omitted).
V.

Construction of Disputed Claims
The parties dispute the meaning of the following five terms and the Court resolves these

disputes as discussed below.
A.

’9944 patent
1.
Term

non-lossy encoding
technique

“non-lossy encoding technique”
Actifio’s Proposed Construction

Delphix’s Proposed Construction

algorithm to create an encoded
representation of data that allows
the original data to be completely
reconstructed

data compression algorithm to create
an encoded representation of data that
allows the original data to be
completely reconstructed from the
encoded data

The parties dispute the meaning of the term “non-lossy encoding technique” in claim 1 of
the ’9944 patent. ’9944 patent, 34:24-27; D. 97 at 12; D. 98 at 10. Claim 1 states, in relevant
part:
A method of storing deduplicated images in which a portion of the image is stored in
encoded form directly in a hash table, the method comprising . . . receiving content to be
included in the deduplicated image of the data object; generating a new hash structure for
the received content; determining if the received content may be encoded using a
predefined non-lossy encoding technique and in which the encoded value would fit
within the field for containing a hash signature; if so, placing the encoding in the field of
the new hash structure instead of placing a hash signature for the received content in the
field, and marking the new hash structure to indicate that the field contains encoded
content for the deduplicated image, so that the received content is not stored since the
received content can be derived from the encoding . . . .

5

’9944 patent, 34:6-8, 21-34. The term is used when describing the storage of a data chunk in the
field of a content “handle” otherwise reserved for a “hash” identifier1, id., 27:55-28:9, where
each unique data chunk is stored only once and associated with a specific handle, D. 98 at 6; D.
97 at 4-5. A data object, such as a text file, is constructed using data chunks as building blocks.
D. 98 at 6. While both parties agree that the term, at the very least, refers to an algorithm
potentially applied to a data chunk, they dispute whether the algorithm (1) is a data compression
algorithm and (2) requires that the original data be decoded from the encoded data itself, rather
than from some other external data. D. 97 at 9; D. 98 at 10.
As to the first aspect of the dispute, Delphix urges the Court to restrict the term to a data
compression algorithm because of the term’s inclusion of “non-lossy” in describing the
“encoding technique.” D. 98 at 10; D. 101 at 7-8. Delphix’s construction would limit the scope
of claim 1 to a method that determines if a compression algorithm can be applied to a data chunk
such that the compressed data chunk can be stored in the content handle itself. This would
exclude determining if an encryption or conversion algorithm, for example, can also be applied
to the data chunk to be stored. Such a limitation, however, is not supported by the claim
language, specification or extrinsic evidence. Notably, the claim language does not reference
data compression and the disputed term is not otherwise defined. Thorner v. Sony Computer
Entm’t Am. LLC, 669 F.3d 1362, 1366-67 (Fed. Cir. 2012) (describing how a patentee must
explicitly redefine a term; otherwise, the plain and ordinary meaning of the term controls).
Based on the parties’ submissions, “encoding,” standing alone, is understood to mean “the
process by which one representation of information is translated into a different representation of

1

A mathematical function creates a unique hash value—an identifier—based on a data
chunk, such that the function would yield the same hash value for two identical data chunks. D.
97 at 7; D. 98 at 6.
6

the same information.” D. 97 at 13; see D. 98 at 12; D. 101 at 6. Likewise, “non-lossy” or
“lossless”—as opposed to “lossy”—is understood to describe the preservation of all the original
information. D. 101 at 7. Put together here, the plain and ordinary meaning of a “non-lossy
encoding technique” is properly understood as a translation process that preserves the original
information, but does not necessarily compress such information. As noted by Delphix and its
expert, data encryption is such a process and it “meet[s] Actifio’s construction.” D. 98 at 12; see
D. 102 at 7. That is, data encryption is understood as a non-lossy encoding technique because it
is a “class of encoding algorithms that create[s] an encoded representation of data that allows the
original data to be completely reconstructed.” D. 98 at 12. While the specification discusses
“lossless [non-lossy] data compression” (brackets added) and “compression,” it does so in
regards to data storage generally, such as in storing data to a storage pool or persistent storage,
’994 patent, 4:38-40, 52-54; 11:62-64; 27:24-31, 40-45, but not in regards to the storage of data
in a field of the content handle itself. In light of this distinction, the Court cannot deem the use
of “non-lossy” as limiting the term solely to data compression algorithms. See, e.g., Thorner,
669 F.3d at 1367-68 (refusing to recognize that an applicant redefined a broad term to mean a
more specific term where the specific term is also used and would otherwise be unnecessary).
To be fair, the single embodiment in the specification, as an example, mentions the use of
“run-length encoding (RLE)” compression in storing data in the content handle. ’9944 patent,
27:63-28:4. As discussed by Delphix, such an embodiment might be the most practical one
where the typical hash field size of the content handle is so small that an uncompressed data
chunk would never fit. D. 101 at 5. It is well established, however, that an embodiment itself
cannot limit the otherwise broad language of a claim. See Thorner, 669 F.3d at 1366 (noting that
a limitation in an embodiment is not enough to constitute disavowal of a claim’s otherwise broad

7

scope). Notably, the claim and specification do not explicitly limit the field size, which could be
set large enough to fit an uncompressed (or encrypted) data chunk. See ’9944 patent, 27:59-64;
34:12-27.

Likewise, the size of a data chunk is not limited and, if set small enough, an

uncompressed (or encrypted) data chunk could fit in the field. See id., 25:61-65; 34:9-20. A
larger field size to accommodate uncompressed data chunks, or smaller uncompressed data
chunks to fit in the field, might not be the most efficient embodiment, but are nonetheless
covered by the plain meaning of the claim. In other words, Delphix has not “shown that [its]
construction must be adopted in order to avoid inoperability.” See Atlas IP, LLC v. St. Jude
Med., Inc., 804 F.3d 1185, 1189-90 (Fed. Cir. 2015).

Accordingly, “non-lossy encoding

technique” in claim 1 cannot be said to refer solely to data compression algorithms and the Court
rejects Delphix’s limitation.
As to the second aspect of the dispute, Delphix proposes the construction of the term
“non-lossy encoding technique” to mean an algorithm that encodes data such that the data can be
completely reconstructed from the encoded data itself. D. 98 at 13-14; D. 101 at 9-10. In
contrast, Actifio argues that such a construction is an inappropriate limitation because the use of
“external inputs” can be used to reconstruct the data that was encoded. D. 97 at 14-15; D. 102 at
9. Claim 1 describes determining if content (a data chunk) can be encoded using a non-lossy
encoding technique and, if so, determining if that encoded content fits within the field of the
content handle otherwise reserved for a hash. ’9944 patent, 34:24-27. If the encoded content
fits, it is placed within the hash field and marked to indicate that the field contains the content
itself “so that the received content . . . can be derived from the encoding.” Id., 34:28-34. Staying
true to the language of claim 1, accord Phillips, 415 F.3d at 1316, and based on the plain
meaning of the terms as discussed above, Delphix’s construction is appropriate because of the

8

use of “non-lossy.” If “non-lossy” means that the original data is preserved in its entirety, the
term properly describes an algorithm that encodes the original data such that it “can be derived
from the encoding,” ’9944 patent, 34:32-35—that is, reconstructed from the encoded data itself.
Otherwise, the algorithm would not be encoding the data in a “non-lossy” form.
While Actifio urges the Court to broaden the scope of the term “non-lossy encoding
technique” to encompass the creation of encoded data that can be reconstructed from some other
external inputs, D. 97 at 14; D. 102 at 9, no such inputs, aside from “the encoding” itself are
mentioned or inferred in the claim or specification, ’9944 patent, 34:34-35. Actifio also argues
that Delphix’s construction renders claim 4 superfluous and forecloses delta compression, an
embodiment disclosed in the specification. D. 97 at 15; D. 102 at 9-10. From Actifio’s
perspective, Delphix’s construction of claim 1 equates the storage of data in the hash field “so
that the received content . . . can be derived from the encoding,” id., 34:32-34; D. 97 at 12; D.
102 at 9-10, with claim 4 that addresses “subsequently reconstituting the content from the
encoded content,” ’9944 patent, 34:45-46. Delphix correctly points out, however, that claim 1
addresses the output of the “non-lossy encoding technique,” the encoded content. D. 101 at 1011. Claim 4, in contrast, describes what happens with the encoded content from claim 1, in that
it is “subsequently” reconstituted. This would include building a data object from the encoded
content (data chunks) as pieces of the object, as well as the use of delta compression. ’9944
patent, 28:15-25; see D. 101 at 10-11. As described by Actifio, delta compression only stores
changed content (data chunks), D. 102 at 5, and, depending on the circumstances, that changed
content (as well as the original content) would first be encoded using a “non-lossy encoding
technique” and then a new version of the data object could be built by “subsequently
reconstituting” the original and changed encoded content, D. 101 at 10-11.

9

At oral argument, Actifio emphasized that, at minimum, a decoding algorithm must be
applied to the encoded content to access the original data and for delta compression to operate
correctly. See Markman Hearing Transcript at 39-40. That is, the decoding algorithm is the
external input required to reconstruct the encoded content. See id. As Delphix pointed out,
claim 1 states that “the received content can be derived from the encoded data,” ’9944 patent,
34:32-34, and the word “derived” implies that some operation is performed on the encoded
content to access the original data. See Hr’g Tr. at 57. That is, a decoding algorithm is not an
external input to the encoded data, but rather an operation applied to the encoded data to access
the original data it contains. Delphix’s construction does not read out the use of a decoding
algorithm necessary for delta compression to work. Accordingly, the second part of Delphix’s
construction is true to the scope of claim 1 and does not render claim 4 superfluous. See, e.g.,
World Class Tech. Corp. v. Ormco Corp., 769 F.3d 1120, 1125-26 (Fed. Cir. 2014) (recognizing
that claim differentiation is not implicated where a construction does not give different claims
the same scope).
2.
Term
encoded value

“encoded value”
Actifio’s Proposed Construction

Delphix’s Proposed Construction

the result of applying the non- the result of applying the non-lossy
lossy encoding technique
encoding technique to the received
content

The parties also dispute the meaning of “encoded value” in claim 1, ’9944 patent, 34:2327; D. 97 at 15; D. 98 at 13, which is used in the same sentence of the ’9944 patent as “non-lossy
encoding technique,” discussed above. At first blush, the parties agree that an “encoded value”
is the output or result of applying the “non-lossy encoding technique,” but they appear to
disagree as to what input the encoding technique is applied. Id. However, in its reply brief,
Actifio agrees that “[t]his language requires the encoding to be applied to ‘the received
10

content,’” but urges the Court to reject Delphix’s construction as “unnecessary and redundant”
because “the plain language of the claim already includes [Delphix’s proposed] limitation.” D.
102 at 11. The Court recognizes that the input from which an “encoded value” is created may be
unclear where claim 1 specifies “received content” as the subject of the disclosed processes, but
“encoded value” otherwise appears with little context.

’9944 patent, 34:21-27 (“receiving

content to be included in the deduplicated image . . . generating a new hash structure for the
received content . . . determining if the received content may be encoded . . . in which the
encoded value [of the received content] would fit within the field for containing the hash
signature”) (emphasis and brackets added). Considering Actifio agrees with the substance of
Delphix’s construction, but opposes its adoption, and to the extent the scope of the term is
otherwise unclear in regards to the input, the Court adopts Delphix’s construction. See O2 Micro
Int’l Ltd. v. Beyond Innovation Tech. Co., 521 F.3d 1351, 1361 (Fed. Cir. 2008) (recognizing
under Markman that district courts are to resolve the parties’ dispute where it goes to “not the
meaning of the words themselves, but the scope that should be encompassed by [the] claim
language”) (emphasis in the original)).
B.

’769 Patent
1.
“identifying the existence of a prior point-in-time image of the application
data in a second set of a plurality of point-in-time images simultaneously stored
at the destination storage pool for a time state prior to the specified restore timestate”
Term

Actifio’s Proposed Construction

Delphix’s Proposed Construction

identifying the existence of No construction necessary; plain finding from two or more point-intime copies of the application data in
a prior point-in-time image and ordinary meaning
the destination storage pool one that is
of the application data in a
before the specified restoration time
second set of a plurality of
point-in-time
images
simultaneously stored at the
destination storage pool for
a time state prior to the

11

specified restore time-state

The parties dispute whether construction is necessary of a term in claim 1 of the ’769
patent addressing a portion of a data restoration operation. ’769 patent, 34:35-40. Delphix urges
the Court to adopt its purportedly simpler construction because, left alone, the term would
otherwise confuse the jury. D. 98 at 17-18.
As discussed by Actifio, the only terms that may be considered “technical” are “storage
pool” (subject to an agreed upon construction, D. 97, Appx. A) and “point-in-time image” (stated
in the claim as “representing the entire application data at a back-up time,” ’769 patent, 33:6466). D. 97 at 16. As such, the Court does not need to clarify these terms further. See Epistar
Corp. v. Int’l Trade Comm’n, 566 F.3d 1321, 1334 (Fed. Cir. 2009) (recognizing the “heavy
presumption that claim terms carry their full ordinary and customary meaning”). Just because a
party may not agree with a word or phrase used in the disputed term, that does not make the term
technical, ambiguous or confusing to a jury, requiring construction. See O2 Micro, 521 F.3d at
1361 (discussing how district courts are to resolve disputes over the scope of a claim). Delphix
specifically takes issue with the use of “identifying” versus “finding,” D. 98 at 17, but to the
extent both parties use the words synonymously, id. (“[t]o identify the existence of something is
to find it”); D. 97 at 7 (“[t]he storage system synchronizes the data in the two storage pools by
finding a common point-in-time image”), the plain and ordinary meaning of “identifying” makes
construction unnecessary. Similarly, the plain and ordinary meaning of “a set of a plurality” also
controls where Delphix concedes that it cannot be “genuinely dispute[d]” that “a set of a
plurality” means “two or more.” D. 98 at 18.
Delphix further argues that construction is necessary because the term “simultaneously
stored” may be misunderstood to mean that each of the point-in-time images were stored at the

12

same time, rather than stored at different times but kept in data storage at the same time. D. 101
at 13. Read in the context of the entire claim, claim 1 discusses how point-in-time images
represent application data at different back-up times. Thus, all point-in-time images cannot be
“simultaneously stored” at the same point in time or they would not reflect changes in the
application data at different times. See ’769 patent, 33:56:67. They would not be back-ups, but
mere copies of the same data at a certain point in time.
Fundamentally, Delphix does not dispute the scope of the term, but merely the words
used. As Delphix notes, its “constructions do not read any limitations into the claim from the
specification . . . they merely clarify for the jury what the claim terms means.” D. 101 at 13. As
recognized by courts in this district, “the Court is not required to provide additional language
construing a claim if its ordinary meaning can be readily understood by a layperson and adopting
it would resolve the parties’ dispute concerning interpretation.” Keurig, Inc. v. JBR, Inc., No.
11-cv-11941-FDS, 2013 WL 1213061, at *6 (D. Mass. Mar. 22, 2013) (citing O2 Micro, 521
F.3d at 1361) (collecting cases). While the term may admittedly be dense, the plain and ordinary
meaning of the words comprising the term, when read together in the context of claim 1, do not
require construction. Id.; see Acumed LLC v. Stryker Corp., 483 F.3d 800, 806 (Fed. Cir. 2007)
(noting that “a sound claim construction need not always purge every shred of ambiguity”).

13

2.
“identify a common point-in-time image between the first set of point-intime images and the second set of point-in-time images”
Term

Actifio’s Proposed Construction

Delphix’s Proposed Construction

identify a common point-in- No construction necessary; plain finding from two or more point-intime image between the and ordinary meaning
time copies of application data in the
first set of point-in-time
destination storage pool and in the
images and the second set
source storage pool one that exists in
of point-in-time images
both storage pools

Delphix urges the Court to construe a term in claim 5 of the ’769 patent—addressing a
portion of the process in which a backup operation is performed, ’769 patent, 35:21-23—for
substantially the same reasons as claim 1. D. 98 at 21. Here, Delphix focuses on the word
“common” as rendering the term ambiguous and confusing, requiring construction.

Id.

Specifically, Delphix’s construction seeks to clarify a “common” point-in-time image to mean
one that exists in both the “destination” and “source” storage pools. Id. However, such a
construction is unnecessary. The plain and ordinary meaning of “common” in relation to pointin-time images, read in the context of claim 5, is understood as a point-in-time image existing in
both sets of point-in-time images. See ’769 patent, 35:1-23. Delphix’s construction is also not
warranted where destination and source storage pools are explicitly used elsewhere in the claims,
but omitted in the disputed term and in claim 5 as a whole. Here, a “set of point-in-time images”
cannot be limited to mean, as Delphix suggests, “point-in-time copies . . . in the
[destination/source] storage pools.”

See Thorner, 669 F.3d at 1367-68 (recognizing that a

“patentee is free to choose a broad term and expect to obtain the full scope of its plain and
ordinary meaning”). As to Delphix’s proposed construction of “identifying” as “finding,” the
Court has already rejected the same argument as to another claim, discussed above, and it is no
more persuasive as to this claim. Accordingly, the Court rejects Delphix’s proposed construction
as to this term.

14

C.

’126 Patent
1.
Term

“performed sequentially at a first time”
Actifio’s Proposed Construction

Delphix’s Proposed Construction

performed sequentially at a No construction necessary; plain performed one after another at a
first time
and ordinary meaning
particular time

The disputed term appears in a portion of claim 1 in the ’126 patent describing the
coordination of “data management functions” (the storage of data to different locations at
different times) to reduce repeated access to the data being stored. ’126 patent, 33:53-58, 34:1115. While conceding that the disputed term initially “appears easily understandably at first
glance,” Delphix argues that the term requires construction because the phrase “at a first time” in
the term otherwise renders the plain and ordinary meaning of “performed sequentially”
ambiguous and confusing. See D. 98 at 24; D. 101 at 14.
While the phrase “at a first time” may seem odd in isolation, “at the first time” is
subsequently used throughout claim 1,’126 patent, 34:15-33 (emphasis added), and gives context
to the disputed term. When read together with the entire claim and specification, “at a first time”
refers to a particular time for the first and second data management functions to be performed
sequentially, and “at the first time” refers back to that time as each of the functions are
performed. That is, the second data management function ends up being dependent on the first
data management function such that they are performed sequentially. For example, as discussed
by Delphix, when two data management functions end up overlapping and require storage of the
same primary data at the same particular time, the first data management function stores the data
to secondary storage to satisfy the function “at the first time,” and the second data management
function—to store the same primary data to backup storage—compares the data stored in
secondary storage to a previous copy in backup storage and only stores the differences to satisfy
15

the second function at that same “first time.” See D. 98 at 23; ’126 patent, 34:11-33. This
process is designed so that “the primary storage is accessed only once for satisfying the first data
management function and the second data management function to be performed sequentially at
the first time.” ’126 patent, 34:30-33 (emphasis added). Accordingly, the plain and ordinary
meaning of the term controls.
VI.

Conclusion
For the foregoing reasons, the disputed claim terms are construed as follows:
1. the term “non-lossy encoding technique” means “algorithm to create an encoded
representation of data that allows the original data to be completely reconstructed
from the encoded data;”
2. the term “encoded value” means “the result of applying the non-lossy encoding
technique to the received content;”
3. the term “identifying the existence of a prior point-in-time image of the application
data in a second set of a plurality of point-in-time images simultaneously stored at the
destination storage pool for a time state prior to the specified restore time-state” does
not require construction;
4. the term “identify a common point-in-time image between the first set of point-intime images and the second set of point-in-time images” does not require
construction, and;
5. the term “performed sequentially at a first time” does not require construction.
So Ordered.
/s/ Denise J. Casper
United States District Judge

16