NOTE: This disposition is nonprecedential. United States Court of Appeals
for the Federal Circuit
CA, INC.,
Appellant
v.
NETFLIX, INC.,
Appellee
2024-1082
Appeal from the United States Patent and Trademark Office, Patent Trial and Appeal Board in No. IPR2022-00322.
Decided: July 29, 2025
DAN YOUNG, Quarles & Brady LLP, Highlands Ranch, CO, argued for appellant. Also represented by KENT DALLOW, MATTHEW CHRISTIAN HOLOHAN.
HARPER BATTS, Sheppard Mullin Richter & Hampton LLP, Menlo Park, CA, argued for appellee. Also represented by JEFFREY LIANG, CHRISTOPHER SCOTT PONDER.
2Before MOORE, Chief Judge, PROST, Circuit Judge, and
HALL, District Judge.
CA, Inc. (“CA”) appeals from a final written decision of the United States Patent Trial and Appeal Board ( “Board”) in an inter partes review (“IPR”). The Board determined that claims 1–20 of U.S. Patent No. 8,656,419 (“the ʼ419 Patent”) are unpatentable as obvious. We affirm.
BACKGROUND
The ’419 Patent is directed to “distributed computing, and more specifically to a dynamic distributed evaluator.” ’419 Patent, col. 1 ll. 5–7. Claims 1, 12, and 18 are independent, and the parties agree that claim 1 is representative of the independent claims. Claim 1 recites:
An apparatus, comprising a first node of a network, the first node comprising:
an interface operable to:
communicate with a second node of the network; and
one or more processors operable to:
tell a plurality of nodes to perform an operation comprising a procedure of an application, the plurality of nodes comprising [the]2 second node and one or more additional nodes;
3instruct the plurality of nodes how to perform the operation using computer code; and tell the plurality of nodes what to do with a result of the operation, and wherein the one or more processors does not know which one of the plurality of nodes will perform the operation.
Id. at claim 1. Dependent claims 7 and 15 additionally require “the code programmed to remove itself upon the completion of a[] process or upon the expiration of a timer.” Id. at claims 7, 15. Dependent claim 8 requires “the first node instructing the second node to remove the code upon completion of a process.” Id. at claim 8. And dependent claims 9 and 16 require “instructing the [second node / plurality of nodes] to use a first code comprising a first programming language and a second code comprising a second programming language.” Id. at claims 9, 16.
In January 2022, Netflix, Inc. (“Netflix”) filed a petition for IPR of the ʼ419 Patent, challenging claims 1–20 as unpatentable for obviousness under 35 U.S.C. § 103. J.A. 55, 66. Netflix asserted that claims 1–4, 6–10, 12, 13, 15, 16, and 18 were obvious in view of the Verbeke3 reference. J.A. 66. Netflix asserted that the remaining dependent claims were obvious in view of Verbeke in combination with other references. Id.
In its final written decision, the Board determined that Netflix “ha[d] demonstrated by a preponderance of the4 evidence that claims 1–20 are unpatentable as obvious under § 103(a).”4 J.A. 41.
CA appeals. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(4)(A).
DISCUSSION
“We review the Board’s legal determination of obviousness de novo and its factual findings for substantial evidence.” Outdry Techs. Corp. v. Geox S.p.A., 859 F.3d 1364, 1367 (Fed. Cir. 2017). “Substantial evidence is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Intel Corp. v. PACT XPP Schweiz AG, 61 F.4th 1373, 1378 (Fed. Cir. 2023) (quoting Novartis AG v. Torrent Pharms. Ltd., 853 F.3d 1316, 1324 (Fed. Cir. 2017)).
CA did not propose or seek a claim construction of any claim terms in the IPR, and it does not do so on appeal. Instead, CA challenges the Board’s findings that Verbeke discloses certain limitations in claim 1, as well as the additional limitations in dependent claims 7–9 and 15–16.
CA first argues that the Board erred in finding that Verbeke discloses the requirement in claim 1 to “tell a plurality of nodes to perform an operation comprising a procedure of an application.” However, Netflix’s expert, Dr. Jagadish, explained in detail why a person of ordinary skill in the art would understand that Verbeke discloses that limitation, which the Board credited. J.A. 16–17, 23; J.A.5 520–28; J.A. 1782.5 CA may disagree with Dr. Jagadish’s opinion, but his testimony constitutes substantial evidence to support the Board’s finding.6 The Board also relied on disclosures in Verbeke, including Figures 1 and 5, and the testimony of CA’s own expert, which provide further supporting evidence. J.A. 20–27.
CA next argues that the Board erred in finding that Verbeke discloses the requirements in claim 1 of “instruct[ing] the plurality of nodes how to perform the operation using computer code” and “tell[ing] the plurality of nodes what to do with a result of the operation.” Here again, the Board credited the testimony of Dr. Jagadish, who explained at length why a person of ordinary skill in the art would understand that Verbeke disclosed both of6 those limitations.7 J.A. 28–32; J.A. 530–47; J.A. 1783. His testimony constitutes substantial evidence supporting the Board’s finding. The Board also relied on disclosures in Verbeke, including Figure 5 and paragraph 82, and the testimony of CA’s own expert, which provide further support. J.A. 28–32.
Finally, CA argues that the Board erred in finding that Verbeke discloses the requirements of dependent claims 7– 9 and 15–16. The Board expressly credited Dr. Jagadish’s expert testimony that “particularly explain[ed]” how Verbeke teaches the additional limitations of claims 7–97 and 15–16.8 J.A. 35–39; J.A. 566–72; J.A. 1784–85. Here, again, that testimony constitutes substantial evidence supporting the Board’s finding that Verbeke discloses those limitations. The Board also relied on disclosures in Verbeke, including paragraphs 140 and 147, and the testimony of CA’s own expert, which provide further evidentiary support. J.A. 35–39.
CONCLUSION
We have considered CA’s remaining arguments and find them unpersuasive. For the foregoing reasons, we affirm the Board’s decision that the challenged claims of the ʼ419 Patent are obvious.
AFFIRMED
Footnotes
- ↩ 1 Honorable Jennifer L. Hall, District Judge, United States District Court for the District of Delaware, sitting by designation.
- ↩ 2 The United States Patent and Trademark Office issued a Certificate of Correction. J.A. 51.
- ↩ 3 U.S. Patent Application Pub. No. 2004/0261069 A1 (“Verbeke”).
- ↩ 4 Because the claims at issue have an effective filing date prior to March 16, 2013, the effective date of the applicable provisions of the Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284 (2011) (“AIA”), the pre-AIA version of 35 U.S.C. § 103 applies.
- ↩ 5 Dr. Jagadish explained, for example, that “Verbeke teaches that a first node (e.g., job submitter 1206 in FIG. 6) tells a plurality of nodes (e.g., worker nodes 1202a-1202h of peer group 1200) to perform an operation (e.g., a task of an application as discussed above) in the distributed system via the task dispatcher node or nodes (and/or a repository node or nodes).” J.A. 17; J.A. 524.
- ↩ 6 CA contends that the Board “misunderst[ood]” the scope of the “tell[ing] a plurality of nodes to perform an operation” limitation. Appellant’s Br. 56. That sounds like a claim construction argument, but CA never asked for a claim construction in the IPR. That may be because, as the Board noted in its decision, CA’s argument in the IPR about the scope of this limitation was “inconsistent with its district court infringement contentions . . . .” J.A. 22 n.3. Regardless of CA’s reasons for not requesting a claim construction in the IPR, CA has not asked for a claim construction on appeal, nor would it have been appropriate to ask for a claim construction for the first time on appeal.
- ↩ 7 Dr. Jagadish explained that “Verbeke repeatedly teaches that its job submitter instructs the framework how to perform the tasks it submits using particular code, either as-identified or as-submitted, or both” and that a person of skill in the art “would have understood this to be instructing a plurality of nodes how to perform an operation . . . because submitting identification of code or code to a framework for distributed processing is instructing or telling a plurality of nodes in the framework how to perform tasks using the code – as that is the entire purpose of submitting tasks/code to a distributed computing system.” J.A. 29; J.A. 537. He further explained that “the identification of the particular tasks . . . and how they are to be performed is defined by [Verbeke’s] job submitter . . . , which results in the particular job repository structure in a job repository node or nodes shown in, for example, FIG. 5 of Verbeke” and that a person of skill in the art “would have recognized that, by virtue of these instructions and the creation of the job repository structure in a job repository node or nodes, the job submitter is also instructing the plurality of nodes what to do with the task results, at least because they are returned to the job repository defined by the job submitter’s instructions.” J.A. 31; J.A. 543.
- ↩ 8 For example, Dr. Jagadish explained that Verbeke discloses the additional limitations of dependent claims 7, 8, and 15 because it “teaches code programmed to remove itself upon the completion of a process or upon the expiration of a timer (e.g., when a job or tasks are completed) and instructing the framework to do the same.” J.A. 35; J.A. 566. Dr. Jagadish also explained that Verbeke discloses the additional limitations of dependent claims 9 and 16 because a person of skill in the art “would have understood by 2008-2009 that distributed systems may run tasks on individual nodes in different programing languages.” J.A. 37, 39; J.A. 571. He further explained that “Verbeke teaches that particular nodes can be part of multiple peer groups,” so a person of skill “would therefore have understood that individual nodes (e.g., a second node) may be part of two different worker peer groups directed to two different aspects of the processing of an application using two different programming languages that would, therefore, perform distributed tasks in different programming languages.” J.A. 37, 39; J.A. 571.