EGENERA INC v. CISCO SYSTEMS INC (2025)

Egenera, Inc. (“Egenera”) appeals from a judgment of the United States District Court for the District of Massachusetts holding that Cisco Systems, Inc. (“Cisco”) did not infringe the asserted claims of Egenera's U.S. Patent No. 7,231,430 (“the ’430 patent”). The district court granted summary judgment of noninfringement as to claims 1 and 5 of the ’430 patent and later, following a jury trial, entered judgment based on the jury's verdict of noninfringement of claims 3 and 7. On appeal, Egenera argues that the district court erred in granting summary judgment with respect to claims 1 and 5 and by denying its post-trial motions for judgment as a matter of law (“JMOL”) or alternatively a new trial on claims 3 and 7. We affirm.

I

The specification of the ’430 patent explains that conventional server systems must be manually deployed or reconfigured to be adapted for a specific use or application. This is often a complicated, time-consuming process requiring a person to physically rewire as many as “100 discrete connections,” each of which “represents a failure point” if rewired incorrectly. ’430 patent at 1:42-45. The ’430 patent seeks to improve upon these prior art systems through the use of a digitalized “processing platform from which virtual systems may be deployed through configuration commands.” Id. at 2:45-47. “The platform provides a large pool of processors from which a subset may be selected and configured through software commands to form a virtualized network of computers” – referred to as a virtualized processing area network – through which an administrator can virtually manage processing resources “through software via configuration commands ․ rather than through physically providing servers, [and] cabling network and storage connections.” Id. at 2:47-62. As a result of this purported innovation, Egenera's server systems need only be physically wired once and then can be reconfigured for specific applications virtually.

Egenera alleges that Cisco's Unified Computing System (“UCS”) infringes claims 1, 3-5, and 7-8 of its ’430 patent.¹ As summarized by the district court, the parties agree that Cisco's UCS is “a scalable computer platform” that includes physical components such as servers, central processing units (“CPUs”), and network interface cards (“NICs”). J.A. 45-46, 78. The UCS virtualizes server management by connecting multiple servers through a single switch and using software, rather than physical cables, to group and deploy servers for specific processing needs. To virtually arrange servers in this manner, UCS uses the NICs to create virtual NICs (“vNICs”) which, in turn, group servers together into virtual local area networks (“VLANs”) using software commands.

Asserted claims 1 and 3 of Egenera's ’430 patent are device claims, while claims 5 and 7 are method claims. Relevant here, claims 1 and 5 require “a plurality of computer processors” and “at least one control node” that emulate Ethernet functionality over an internal communication network. Claims 3 and 7 require processors to be programed to establish a specified virtual local area network topology.² Below we reproduce the pertinent portions of the claims that are at issue in this appeal:

1. A platform for automatically deploying at least one virtual processing area network, in response to software commands, said platform comprising:

a plurality of computer processors connected to an internal communication network;

at least one control node in communication with an external communication network and in communication with an external storage network having an external storage address space, ․;

․

wherein the plurality of computer processors and the at least one control node include network emulation logic to emulate Ethernet functionality over the internal communication network.

3. A platform for automatically deploying at least one virtual processing area network, in response to software commands, said platform comprising:

a plurality of computer processors connected to an internal communication network;

․

configuration logic for ․ a virtual local area network topology defining interconnectivity and switching functionality among the specified processors of the virtual processing area network ․ said configuration logic including logic to select, under programmatic control, a corresponding set of computer processors from a plurality of computer processors, to program said corresponding set of computer processors and the internal communication network to establish the specified virtual local area network topology ․

5. A method of automatically deploying at least one virtual processing area network, in response to software commands, said method comprising the acts of:

providing a platform having a plurality of computer processors and at least one control node connected to an internal communication network ․;

․

wherein the plurality of computer processors and the at least one control node emulate Ethernet functionality over the internal communication network.

7. A method of automatically deploying at least one virtual processing area network, in response to software commands, said method comprising the acts of:

providing a platform having a plurality of computer processors and at least one control node connected to an internal communication network, wherein the at least one control node is in communication with an external communication network and an external storage network having an external storage address space;

receiving software commands specifying ․ a virtual local area network topology defining interconnectivity and switching functionality among the specified processors of the virtual processing area network ․;

․

under programmatic control and in response to the software commands, programming said corresponding set of computer processor[s]; and the internal communication network to establish the specified virtual local area network topology providing communication among said corresponding set of computer processors but excluding the processors from the plurality not in said set ․

During the claim construction process, the parties informed the district court they had disputes over the proper construction of several claim terms. Among the disputed terms were “computer processor/processor” and “emulate Ethernet functionality over the internal communication network,” as used in claims 1 and 5.

With respect to “computer processor/processor,” the court rejected Egenera's construction, which would have defined the term as referring to a “processing node” described in the specification of the ’430 patent, thereby encompassing the UCS NICs. The court, instead, decided to accord “computer processor” its ordinary meaning of, simply, “a CPU.” J.A. 1239-44.

As to “emulate Ethernet functionality over the internal communication network,” Cisco argued that “because the Ethernet functionality is ‘emulate[d],’ it is necessarily absent from the internal communication network.” J.A. 1259 (alteration in original). The district court rejected Cisco's proposed construction, which it viewed as improperly “narrow[ing] the scope of the ‘internal communication network,’ ” finding instead that the specification describes the platform as “agnostic as to the specific architecture of the internal communication network.” J.A. 1260. In its claim construction order, the district court addressed only the portion of the “emulate” term that the parties disputed, i.e., whether “emulate” implies absence from the internal communication network; it did not also determine what the patent specifically means by “emulate Ethernet functionality,” presumably because neither party advised the court that this point was disputed and could be resolved through claim construction.³

Following claim construction, Cisco moved for summary judgment of noninfringement as to all four claims. The district court granted the motion with respect to claims 1 and 5 and denied it with respect to claims 3 and 7. As to claims 1 and 5, the district court agreed with Cisco that, given its construction of “computer processor/processor” as a CPU, the record could not support a reasonable finding that Cisco's UCS contains a “plurality of computer processors [that] include network emulation logic to emulate Ethernet functionality over the internal communication network” because the CPUs do not include the NICs. J.A. 48-49. The district court first held the undisputed evidence demonstrated that “Ethernet emulation functionality resides with [vNICs] and interfaces – stand-alone components separate and apart from the CPUs,” and therefore this functionality did not meet the claim requirement that emulation occur at the CPUs. J.A. 48. The Court then rejected Egenera's argument that Cisco's UCS CPUs’ undisputed communication on, and use of, the internal communication network amounted to emulation of Ethernet functionality, explaining that “knowledge and use of a communications network is not emulation of the functionality of that network.” J.A. 49.

At a subsequent trial, a jury found Egenera failed to prove that Cisco infringes claims 3 and 7 of the ’430 patent. Egenera then moved for JMOL of infringement or, in the alternative, for a new trial. The district court denied both motions.

In denying Egenera's JMOL motion, the district court explained that the record “provides ample basis to support the jury's findings under the plain language of the claims,” J.A. 78, adding “the fact that the jury did not weigh the evidence as Egenera might have wished does not mean that the jury improperly imported additional limitations from outside the claim language,” J.A. 79. The district court did not find it necessary to expressly address every contention Egenera raised. Instead, after demonstrating how a reasonable jury could have found Egenera failed to prove Cisco's UCS practiced the “topology” limitation as required by claims 3 and 7, the court determined this was a sufficient failing to deprive Egenera of any entitlement to a judgment of infringement. J.A. 78.

The district court then proceeded to reject each of Egenera's proffered grounds for a new trial. It concluded that Cisco's alleged violations of in limine orders, about which Egenera had failed to raise objections during trial, did not constitute plain error. J.A. 79-81. The court also held that Egenera waived its challenge to purportedly improper testimony and, regardless, had not shown it was prejudiced by the testimony. J.A. 81-82. Next, the district court rejected Egenera's contention that an instruction stating patented products can still infringe other patents “was necessary to ensure the jury did not robotically assume noninfringement.” J.A. 82-83. Lastly, the district court disagreed with Egenera that a curative instruction was necessary to alleviate error in an introductory remark the court had made to prospective jurors during jury selection. J.A. 83

Egenera timely appealed. We have jurisdiction pursuant to 28 U.S.C. § 1295(a)(1).

II

When handling appeals from a district court, we review matters not unique to patent law according to the law of the regional circuit, here the First Circuit. See AbbVie Deutschland GmbH & Co., KG v. Janssen Biotech, Inc., 759 F.3d 1285, 1295 (Fed. Cir. 2014). Under First Circuit law, we review a district court's grant of summary judgment de novo and will affirm if, when taking the evidence in the light most favorable to the non-moving party, “there is no genuine issue as to any material fact.” Santiago-Diaz v. Rivera-Rivera, 793 F.3d 195, 199 (1st Cir. 2015). The First Circuit reviews a district court's ruling on a motion for judgment as a matter of law under the same de novo standard. See Crowley v. L.L. Bean, Inc., 303 F.3d 387, 393 (1st Cir. 2002). We may reverse the denial of such a motion only if, when viewing the evidence in the light most favorable to the non-movant and drawing all reasonable inferences in that party's favor, “there was only one conclusion the jury could have reached ․ and that result was contrary to the verdict.” Id. (internal quotation marks and citations omitted).

We review the district court's denial of a motion for a new trial for abuse of discretion. See AbbVie Deutschland, 759 F.3d at 1295, 1302 (applying First Circuit law). The same is true of the district court's evidentiary rulings, which we may reverse only for an abuse of discretion. See Neece v. City of Chicopee, 106 F.4th 83, 93 (1st Cir. 2024) (holding timely objections to evidentiary rulings are reviewed for abuse of discretion); Fusco v. Gen. Motors Corp., 11 F.3d 259, 262 (1st Cir. 1993) (holding that district court's determination that evidentiary objection was waived is reviewed for abuse of discretion).

Our review of patent law matters is governed by our own law. See Genetic Veterinary Scis., Inc. v. LABOKLIN GmbH & Co. KG, 933 F.3d 1302, 1315 (Fed. Cir. 2019). “A determination of infringement is a question of fact that is reviewed for substantial evidence when tried to a jury.” ACCO Brands, Inc. v. ABA Locks Mfrs. Co., 501 F.3d 1307, 1311 (Fed. Cir. 2007). Factual findings are “supported by substantial evidence if a reasonable jury could have found in favor of the prevailing party in light of the evidence presented at trial.” Amgen Inc. v. Hospira, Inc., 944 F.3d 1327, 1335 (Fed. Cir. 2019).

III

Egenera appeals the district court's: (i) grant of Cisco's motion for summary judgment of noninfringement of claims 1 and 5; (ii) denial of Egenera's motion for JMOL of infringement of claims 3 and 7; and (iii) denial of Egenera's motion for, in the alternative, a new trial on infringement of claims 3 and 7.

A

The district court granted summary judgment of noninfringement of claims 1 and 5 based on its determination that Egenera failed to raise a genuine dispute of material fact as to whether the CPUs in Cisco's accused UCS product “emulate” Ethernet functionality. In making this determination, the district court followed the parties’ lead and viewed the summary judgment motion as posing solely the question of whether Egenera had produced sufficient evidence from which a verdict of infringement could be sustained. Neither party expressed to the court that the summary judgment motion implicated any dispute as to claim construction and, accordingly, the district court did not further construe any claim terms in resolving the motion.

Egenera now appeals the district court's conclusion that the evidence could not demonstrate, to a reasonable factfinder, that Cisco's UCS practices the “wherein the plurality of computer processors ․ include network emulation logic to emulate Ethernet functionality over the internal communication network” limitation. ’430 patent at 29:35-38 (emphasis added). The district court granted summary judgment based on its conclusion that the record evidence could only support a finding that in Cisco's accused UCS system, the computer processors (i.e., CPUs) merely used Ethernet functionality. J.A. 49. It determined, therefore, that no reasonable factfinder could find that the UCS’ CPUs themselves emulated Ethernet functionality. J.A. 49. It further found that Egenera's evidence showed, at best, that the component of the UCS system that actually “emulates” Ethernet functionality is the network interface card (“NIC”), which is not a computer processor. J.A. 49; see also id. (“[K]nowledge and use of a communications network is not emulation of the functionality of that network ․”) (emphasis added).

If we view this appeal as presenting the question of whether Egenera's evidence before the district court was sufficient to create a genuine dispute of material fact – which, importantly, and as we explain further below, is how Egenera characterizes its appeal – there is little doubt that the district court should be affirmed. A reasonable jury, viewing the evidence in the light most favorable to Egenera, could find only that the UCS CPUs have knowledge of, and use, Ethernet functionality, but do not – as the claims require – actually “emulate” Ethernet functionality.

Egenera's efforts to persuade us otherwise fail. Egenera faults the district court for purportedly “ignor[ing] ․ evidence that the CPUs include Ethernet emulation functionality” and that “the UCS operates consistently with the embodiments of the ’430 Patent itself.” Open. Br. at 33-35. But Egenera's supporting citations are limited to its summary judgment briefing and its patent specification. See id. (citing J.A. 103 (’430 patent at 3:55-63), 5222-24 (Egenera's summary judgment briefing)). It fails to cite any actual evidence the district court overlooked that somehow shows the processors in Cisco's accused UCS system function like any disclosed embodiments in the patent itself.

What we have said to this point fairly deals with the appeal as Egenera has presented it to us. But our review of this case has raised concern that the proper analysis may not necessarily be as straightforward as what we have set out. It has come to appear on appeal that what the parties were truly disagreeing about at summary judgment was a matter of claim construction, not the sufficiency of Egenera's evidence. In particular, with the benefit of hindsight and further consideration on appeal, one might say that the dispute at summary judgment was actually about the proper construction of “emulate,” and whether, as used in the asserted claims, it encompasses the CPUs’ mere “use” of Ethernet functionality.

While we identified this issue and discussed it with the parties at oral argument, see Oral Arg. at 1:10-5:20, we need not resolve what to do in a case like this, where the parties present a motion for summary judgment to the district court as one involving issues of fact but it becomes clear on appeal that the motion actually turned on an unresolved issue of claim construction. This is because we will not address a claim construction issue for the first time when the appellant has failed to preserve that issue in the district court and has also failed to adequately present it to us for decision on appeal. Here, Egenera did not ask the district court to construe the meaning of either “emulate Ethernet functionality” or “emulate” and, likewise, has not asked us to do so.

As we have already noted, the parties’ Markman briefing identified disputes with respect to the “emulate Ethernet functionality” limitation but only in relation to the proper construction of “computer processor” and “over the internal communication network.” See J.A. 1259-60 (court rejecting Cisco's proposal to limit “internal communication network” to “non-Ethernet physical network”). Neither Egenera nor Cisco asked the district court to determine what it means to “emulate Ethernet functionality.” The district court resolved the claim construction disputes presented to it but, quite reasonably, did not address disputes that were not presented to it.

Nor did either party identify a claim construction dispute in connection with Cisco's summary judgment motion. In its motion, Cisco argued that no reasonable juror could find it infringed claims 1 and 5 of the ’430 patent because Egenera had failed to produce sufficient evidence to show that the CPUs in the accused UCS system “emulate Ethernet functionality.” No. 1:16-cv-11613-RGS D.I. 140 at 15-18 (Cisco's opening brief in support of its summary judgment motion). Egenera countered that it had presented sufficient evidence to create a genuine dispute of material fact as to whether “[t]he UCS Server CPU communications on (and uses of) these virtual interfaces,” i.e., the interfaces between the CPUs and the UCS Fabric Interconnects, “satisfy the CPU-related Ethernet emulation functionality required in the claims.” J.A. 5222. Egenera expressly denied the existence of any claim construction dispute, insisting, “[t]he only apparent dispute is factual: whether the UCS Server CPUs include network emulation logic to emulate Ethernet functionality.” J.A. 5222 (emphasis added). Responding in like kind in its reply brief, Cisco also treated its motion as solely about Egenera's evidence, and not about claim scope, writing: “Egenera's effort to change the scope of the claim does not present a factual dispute that precludes summary judgment.” No. 1:16-cv-11613-RGS D.I. 189 at 3 (emphasis omitted); see also id. (“Claims 1 and 5 plainly require the emulation logic (and functionality) to be on (or by) the CPUs, and there is no dispute that the CPUs do not have that logic.”) (emphasis omitted). And again, entirely understandably, the district court limited its summary judgment analysis to the issue the parties presented – one of sufficiency of evidence – and did not independently evaluate whether it had to undertake further claim construction. See J.A. 48-49.

Consistent with its approach in the trial court, Egenera has framed its appeal as presenting only a sufficiency-of-the-evidence issue, and not an issue of claim construction. See, e.g., Open. Br. at 32 (quoting J.A. 49) (“[T]he district court made an improper factual finding that ‘knowledge and use of a communications network is not emulation of the functionality of that network.’ ”);⁴ Reply Br. at 3 (“The assessment of fact [i.e., concluding that knowledge and use is not emulation], followed by a conclusion of noninfringement[,] is nothing short of a judicial factual determination at the summary judgment stage – not allowed.”). Egenera makes a fleeting reference to claim construction in its Opening Brief. See Open. Br. at 35 (“The district court's ruling effectively excluding this functionality [i.e., “Ether-net emulation being executed both by CPUs and control nodes”] from the scope of claims 1 and 5 was thus legal error, as ‘[a] claim construction that excludes a preferred embodiment is rarely, if ever, correct’ ․ The district court committed legal error in improperly construing the scope of claims 1 and 5 to exclude a preferred embodiment, and improperly resolved a factual issue in deciding that the UCS does not meet the emulated Ethernet limitations.”). Yet when Cisco suggested in response that Egenera is now arguing claim construction, see Resp. Br. at 27, Egenera explicitly denied doing so, insisting in reply that “Cisco is incorrect that Egenera here seeks a new construction of ‘emulate Ethernet functionality’ ․ The meaning of the word ‘emulate’ is not here the issue,” Reply Br. at 3 (emphasis added). And when we asked at oral argument whether Egenera had appealed the lack of a construction of “emulate,” Egenera told us it had not. See Oral Arg. at 4:53-4:59; see also id. at 18:10-18:30 (Cisco urging us not to engage in sua sponte claim construction).

An appellant “waive[s] the ability to argue for an alternative claim construction” where, as here, it merely “allude[s] to the possibility” that the district court's construction was wrong, “d[oes] not raise the issue in the ‘Statement of the Issues,’ cite[s] no legal support for its claim construction ‘arguments,’ and d[oes] not even recite the [legal] standard of review for claim construction.” Whitserve, LLC v. Comp. Packages, Inc., 694 F.3d 10, 19 n.4 (Fed. Cir. 2012). That is the situation we confront here. Egenera has not, even on appeal, argued that claim construction is necessary, let alone presented a meaningful claim construction analysis of “emulate.” This claim term would call for considerable analysis, and this is not a case where we will undertake the task on our own. See generally Greenlaw v. United States, 554 U.S. 237, 243-44, 128 S.Ct. 2559, 171 L.Ed.2d 399 (2008) (discussing party presentation principle and rationale supporting it).

Accordingly, we affirm the district court's grant of summary judgment of noninfringement of claims 1 and 5 of the ’430 patent.

B

We turn next to Egenera's challenge to the district court's handling of claims 3 and 7. Egenera insists that the court erred in denying its motion for JMOL and entering judgment of noninfringement of claims 3 and 7. We disagree.

At trial, Cisco presented the jury with three alternative grounds for rejecting Egenera's infringement contentions and returning a verdict of noninfringement: its UCS does not practice (i) the network topology limitation, (ii) the modifying messages limitation, and (iii) the extracting and identifying messages limitation. The jury returned a general verdict of noninfringement, making it impossible to know which one or more of these grounds persuaded the jurors. We will uphold such a general verdict “if there was sufficient evidence to support any of the [prevailing party's] alternative factual theories; we assume the jury considered all the evidence” and applied the appropriate burden of proof to all factual theories presented. i4i Ltd. P'ship v. Microsoft Corp., 598 F.3d 831, 849 (Fed. Cir. 2010). Therefore, here, we must uphold the judgment of noninfringement unless Egenera can show that the jury lacked substantial evidence with respect to all three of the grounds on which it may have found noninfringement. We therefore need only address one of Cisco's grounds for noninfringement – whether its UCS practices the network topology limitation – to determine that Egenera has failed to make such a showing.

The topology limitation requires that the processors “establish the specified virtual local area network topology providing communication among said corresponding set of computer processors.” ’430 patent at 32:22-24 (claim 7). The evidence presented at trial was sufficient to support a finding that the UCS establishes network topology at the NICs, but not at the processors, thereby failing to satisfy the network topology limitation of the claims. See, e.g., J.A 12968 (“[B]ecause Cisco is making networks of servers and not processors, they don't program the processors to establish the network topology.”); J.A. 12982 (“Cisco does not need to and does not program the CPUs.”); J.A. 13204 (“The programming of the VNIC ․ establishes the virtual local area topology specified in the claim.”). The jury was free to credit Cisco's evidence on this point and to reject Egenera's contrary contention that the drivers or MAC addresses loaded into the UCS CPUs constitute “programming” meeting the claim limitation. Drawing all reasonable inferences in favor of Cisco, as the verdict winner, the record contains substantial evidence from which the jury could have reasonably found that Egenera failed to prove that the accused UCS meets the network topology limitation.

Egenera's only remaining argument relating to claims 3 and 7 is that the district court purportedly erred by reading an intent requirement into the claims. In denying Egenera's JMOL motion, the court said: “Cisco witnesses testified that the central processing unit ․ is not programmed for the claimed purpose of establishing the specified virtual local area topology.” J.A. 78 (emphasis omitted). The court's statement refers only to the functionality recited in the claims and does not impose an additional intent requirement. Thus, the court did not erroneously ground its upholding of the noninfringement verdict on Egenera's failure to prove a nonexistent intent requirement.

Accordingly, we affirm the district court's denial of Egenera's JMOL motion.

C

Finally, Egenera contends that several alleged errors should have compelled the district court to order a new trial. Applying First Circuit law, we review the trial court's determinations for abuse of discretion. See Borras v. Sea-Land Serv., Inc., 586 F.2d 881, 887 (1st Cir. 1978). A new trial must be granted if “the verdict is against the clear weight of the evidence, or is based upon evidence which is false, or will result in a miscarriage of justice.” Id. at 886 (internal quotation marks omitted). Egenera challenges the district court's rejections of five grounds it raised for a new trial: (i) the verdict is against the weight of the evidence; (ii) the district court should have provided a curative instruction to fix statements made during empanelment; (iii) there were errors in the jury instructions; (iv) Cisco introduced expert opinion from lay witnesses; and (v) Cisco's counsel made improper statements during closing. We find no abuse of discretion in any of these rulings.

1

Egenera contends that a new trial is warranted because the overwhelming weight of the evidence supported a finding of infringement, but the jury nonetheless returned a verdict of noninfringement. Egenera's assessment of the record relies on its denigration of much of Cisco's evidence as irrelevant because, according to Egenera, this evidence compared the accused UCS product with Egenera's patent-practicing product, the “BladeFrame,” rather than properly comparing the UCS to the asserted claims. This is not a fair characterization of the record. Egenera overlooks the impact of its claim of willful infringement, which – by Egenera's own choice – made relevant a body of evidence that might otherwise not have been, including Egenera's allegations of copying and Cisco's responses to that charge. See, e.g., J.A. 12103-04 (Egenera accusing Cisco of copying its product); see also C.R. Bard Inc. v. AngioDynamics, Inc., 979 F.3d 1372, 1380 (Fed. Cir. 2020) (evidence of copying may be relevant to willful infringement).

In addition to Cisco's evidence not being irrelevant, the district court clearly and correctly instructed the jury that infringement is based on a comparison between the patent claims and the accused product. J.A. 15047 (“When deciding the issue of infringement, you must compare Cisco's accused UCS product and its methods of use to the asserted claims of the ’430 patent.”). Hence, not only was the weight of the evidence not overwhelmingly in favor of infringement, but there was also no error in the pertinent jury instructions. We affirm the district court's denial of this ground for Egenera's motion.

2

During the jury selection process, the district court gave the panel of prospective jurors “the basic idea of the case,” with the judge emphasizing he was “going to be brief, nothing like the full explanation you are going to be getting shortly from the lawyers.” J.A. 13444. Then the court explained that “[t]o ‘infringe’ in the patent context means to copy essentially without permission.” Id. Egenera voiced no objection or complaint about this statement until 10 days later, when it asked that the judge tell the jury that copying is not a required element of infringement. By the time Egenera made its request for a curative instruction, all of the evidence and closing argument had been presented to the jury. Additionally, in its final instructions, the district court thoroughly and correctly told the jury that “[w]hen deciding the issue of infringement, you must compare Cisco's accused UCS product and its methods of use to the asserted claims of the ’430 patent.” J.A. 15047; see also J.A. 15048-52 (providing further instructions on infringement). In refusing to give Egenera's curative instruction and in rejecting the associated request for a new trial, the district court acted within its discretion. See United States v. Amador-Huggins, 799 F.3d 124, 131 (1st Cir. 2015) (affirming denial of request for curative instruction made two days after “minor evidentiary issue stemming from an isolated comment”); see also Romano v. U-Haul Int'l, 233 F.3d 655, 667 (1st Cir. 2000) (“A new trial is necessary only if the error could have affected the result of the jury's deliberation.”) (internal quotation marks omitted).

3

Egenera next faults the district court's rejection of its proposed final instruction that would have told the jury “[a] product that is covered by a subsequent or later patent may still infringe an earlier patent.” J.A. 11456. The First Circuit “review[s] jury instructions de novo, recognizing that properly preserved objections to the omission of desired jury instructions constitute reversible error only if the omitted instructions were (1) correct as a matter of substantive law, (2) not substantially covered in the charge as a whole, and (3) integral to an important point in the case.” Cerqueira v. Am. Airlines, Inc., 520 F.3d 1, 15 (1st Cir. 2008). As we already noted, “[a] new trial is necessary only if the error [in the instructions] could have affected the result of the jury's deliberation.” Romano, 233 F.3d at 667 (internal quotation marks omitted).

Egenera has failed to show that the court's refusal to give its proposed instruction was prejudicial, that is, integral to an important issue in the case. The final jury instructions were thorough, accurate, and sufficient to prevent the potential confusion about which Egenera was concerned.⁵ The court told the jury:

Patent law gives the owner of a valid patent the right to exclude others from importing, making, using, offering to sell, or selling the patented device or using the patented method within the United States during the term of the patent. Any person who engages in any of these acts without the patent owner's permission infringes the patent. When deciding the issue of infringement, you must compare Cisco's accused UCS product and its methods of use to the asserted claims of the ’430 patent.

J.A. 15047; see also J.A. 15048-52 (providing more detailed instructions on infringement). There is no basis to conclude that a reasonable juror would have decided the case differently had Egenera's proposed additional instruction been given.

4

Egenera next contends a new trial is warranted because Cisco elicited improper expert testimony concerning copying and infringement from two lay witnesses. The district court held that “Egenera waived this argument by failing to raise any relevant objection during the examination of either witness.” J.A. 81. Egenera responds on appeal by pointing to three instances in which, in its view, it preserved this issue.

Egenera argues that it preserved the issue by filing a pretrial motion in limine. Egenera had moved to “preclude Cisco from offering any lay witness opinion testimony attempting to opine on noninfringement of the accused products or otherwise comparing the accused products to the patent specification or claims.” J.A. 10942 (internal capitalization altered). The district court denied the motion because “the opinion testimony is relevant to intent and knowledge for purposes of defending the willful infringement and indirect infringement claims.” J.A. 165.

Egenera contends it further preserved the issue by moving during trial “to exclude [lay witness Mr. Jayakrishnan] from offering testimony or argument at trial on Exhibit 23 to Egenera's expert Dr. Jones’ infringement report.” J.A. 11775. The court found that this “objection was not raised within the prescribed time,” agreeing with Cisco “that Dr. Jones having challenged Mr. Jayakrishnan's earlier testimony, the door is open for [Mr. Jayakrishnan] to respond.” J.A. 173; see also J.A. 12776 (“My notes show that he was called an adverse witness by Egenera ․ [and] was asked his opinion about infringement, therefore, it's fair that he can reply to Dr. Jones's criticism of the opinions that he gave at Egenera[’s] invitation.”).

Egenera additionally insists it preserved the issue by objecting, pursuant to Federal Rule of Evidence 702, to testimony from Cisco employee Mr. Dvorkin, when he was asked: “Is programming the UCS NIC to establish the network topology the same as programming the UCS server CPU to establish a network topology?” J.A. 12815-16. The district court overruled this objection, explaining that Cisco was “asking [Mr. Dvorkin] about the different approaches and whether they provided the same design.” J.A. 12816.

We discern no abuse of discretion in the district court's determination that neither Egenera's broad pretrial motion in limine, nor its two highly-specific trial objections, were sufficient to preserve the argument that the lay witness testimony of two Cisco employees so prejudiced Egenera that it warrants a new trial. The First Circuit recognizes that, to be fair to the district court, it may be necessary for a party whose pretrial motion to exclude was denied “to renew [its] objection when the evidence is actually offered.” Fusco, 11 F.3d at 262 (“[R]equiring this renewal gives the trial judge a chance to reconsider the ruling with the concrete evidence presented in the actual context of the trial.”). Many in limine orders, including those at issue here, are not self-enforcing; often a party must raise specific objections when it believes such orders are being violated at trial. Here, Egenera's pretrial motion in limine was insufficient to put the district court on notice that Egenera had a preexisting challenge to the specific questions and answers presented at trial. The district court based its in limine ruling on a broad rationale: the potential relevance of the testimony to a willful infringement theory. The decision did not address other potentially relevant objections to the testimony that would arise in the context of trial. The objections Egenera now lodges are too specific and context-dependent to have been conclusively settled by a pretrial motion in limine. And the two highly specific trial objections do not sweep broadly enough to have put the district court on notice of the scope of review Egenera now requests: a searching inquiry into whether the testimony as a whole amounted to improper lay opinion. Accordingly, First Circuit law provides the district court discretion to require Egenera to renew its objections, and to find forfeiture as a consequence of Egenera having failed to do so. See Fusco, 11 F.3d at 262 (explaining that, under First Circuit law, the district court could require a renewed objection to evidence at trial after “the in limine motion to exclude was denied”).

5

Lastly, Egenera asserts it should be granted a new trial because Cisco's closing arguments violated multiple pretrial orders, including those prohibiting “testimony, argument, or reference to the absence of any witness who does not appear at trial;” “empty chair arguments;” “referring to Egenera as a nonpracticing entity or a patent troll;” and any reference to “either Egenera's or Cisco's ability to finance the present litigation, including through litigation funding or by private equity funders.” J.A. 13414.

Because Egenera failed to object to Cisco's closing arguments, either during the argument or in a timely manner after its conclusion, Egenera was required to show that permitting Cisco's closing arguments was plain error. Although Cisco's arguments traveled close to the lines the district court set, we see no abuse of discretion in the district court's conclusion that Cisco never violated the literal terms of its orders.

For example, Cisco referred to Egenera as “venture capitalists trying to make good on their bad bet.” J.A. 13416, 14987. This comes close to describing Egenera as a nonpracticing entity and at least equally close to commenting on Egenera's ability to finance the present litigation. But, strictly speaking, counsel's comments did not directly tell the jury that Egenera is a nonpracticing entity, nor how it is that Egenera was financing its case expenses. The district court did not abuse its discretion in finding that allowance of the closing argument, to which Egenera did not object, did not constitute plain error.

IV

We have considered Egenera's remaining arguments and find them unpersuasive. For the foregoing reasons, we affirm the district court's grant of Cisco's motion for summary judgment of noninfringement of claims 1 and 5 as well as its denial of Egenera's motions for judgment as a matter of law or a new trial with respect to claims 3 and 7.

AFFIRMED

FOOTNOTES

1.   Claims 4 and 8 are not involved in this appeal.

2.   “Network topology” means, in general, the “geography of the network” or the “geometric arrangement of links and nodes of a network.” Network Topology, Newton's Telecom Dictionary (24th ed. 2008).

3.   After concluding the claim construction process, the district court conducted a bench trial and found the ’430 patent invalid due to the failure to name all inventors. On appeal, we affirmed the court's claim constructions but vacated its invalidity finding, remanding for the further proceedings that have now generated the current appeal. See Egenera, Inc. v. Cisco Sys., Inc., 972 F.3d 1367 (Fed. Cir. 2020).

4.   We need not decide whether, had Egenera plainly and clearly made a claim construction argument in its opening brief on appeal, we would have considered it. See generally Kaufman v. Microsoft Corp., 34 F.4th 1360, 1369 (Fed. Cir. 2022) (internal quotation marks omitted) (“[Where] there is no indication that the [district] court was aware of the supposed [claim construction] dispute, a party is considered to have forfeited [it] ․ and cannot resurrect its argument on appeal by pointing to ambiguous statements in the record.”); see also Wash World Inc. v. Belanger Inc., 131 F.4th 1360, 1368 (Fed. Cir. 2025) (holding appellant's claim construction position was forfeited where it gave district court “no opportunity to evaluate [its] position”).

5.   It is undisputed that the district court's instruction was substantively correct, and that Egenera's additional proposed instruction was also substantively correct. Accordingly, we have no occasion to apply our own legal standards to the narrow dispute over the district court's decision to omit Egenera's preferred language. See generally TecSec, Inc. v. Adobe Inc., 978 F.3d 1278, 1290-91 (Fed. Cir. 2020) (reviewing patent law instruction for legal error under Federal Circuit law while assessing prejudicial effect of uncontested instruction according to regional circuit law). The parties are in agreement that the issue on appeal is governed by First Circuit law. See Open. Br. at 30, 50-52; Resp. Br. at 23-25, 47-53; Reply Br. at 13-24.

Stark, Circuit Judge.