Netflix Inc v. Broadcom Inc

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Netflix, Inc. (Delaware)
  • Defendant: Broadcom Inc. (Delaware), VMware LLC
  • Plaintiff’s Counsel: Baker Botts LLP.
• Case Identification: 3:25-cv-03738, N.D. Cal., 11/17/2025
• Venue Allegations: Plaintiff alleges venue is proper in the Northern District of California because Defendants maintain principal places of business in the district, reside there, and have committed acts of patent infringement within the district.
• Core Dispute: Plaintiff alleges that Defendants’ virtualization and network switching products, including technology acquired through Broadcom’s merger with VMware, infringe five U.S. patents related to virtual machine service availability, network subnet provisioning, and high-precision time synchronization in computer networks.
• Technical Context: The patents-in-suit relate to technologies for managing computing resources in virtualized environments and ensuring precise clock synchronization across networked devices, which are foundational to modern cloud computing, data centers, and industrial automation systems.
• Key Procedural History: The complaint notes that on October 29, 2025, the Court issued an order finding asserted claims of all five patents-in-suit were directed to patent-ineligible subject matter under 35 U.S.C. § 101. Plaintiff states it is re-pleading its infringement claims to avoid waiver and preserve its right to appeal the Court's patent eligibility ruling. The complaint also alleges a history of willful infringement by VMware, citing findings from prior litigation.

Case Timeline

Date	Event
2003-03-17	’102 Patent Priority Date
2005-04-27	’912 Patent Priority Date
2005-12-09	’931 & ’751 Patents Priority Date
2007-12-25	’102 Patent Issue Date
2008-11-04	’931 Patent Issue Date
2010-01-19	’912 Patent Issue Date
2010-02-02	’751 Patent Issue Date
2014-08-29	’472 Patent Priority Date
2019-06-25	’472 Patent Issue Date
2022-05-26	Broadcom and VMware enter Merger Agreement
2023-11-22	Broadcom completes acquisition of VMware
2024-12-23	Netflix sends first notice letter regarding infringement
2025-04-15	Netflix sends second notice letter regarding infringement
2025-10-29	Court issues order finding asserted claims patent-ineligible
2025-11-17	Second Amended Complaint Filing Date

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 10,331,472 - “Virtual Machine Service Availability,” issued June 25, 2019

The Invention Explained

• Problem Addressed: The patent describes a problem in virtualized server clusters where ensuring high availability for critical services relied on redundant standby systems. If a primary server failed, the service would operate without a redundant copy during the server's downtime, creating a "risky period of lower service availability" that was undesirable for "real-time or critical services" (’472 Patent, col. 1:54-61). The conventional approach of synchronous replication to maintain redundancy was described as negatively affecting both performance and cost efficiency (’472 Patent, col. 1:50-51).
• The Patented Solution: The invention proposes a system and method where a "service availability controller" monitors the availability of various services running on virtual machines (’472 Patent, col. 2:58-63). If availability for a high-priority service is reduced, the system automatically creates capacity for it by deactivating an instance of a lower-priority service on a different server. A new virtual machine is then activated on that freed-up server to run an instance of the high-priority service, thereby restoring redundancy and improving reliability without requiring additional hardware costs (’472 Patent, Abstract; col. 1:65-2:8).
• Technical Importance: The technology provided a method for dynamically reallocating software-based resources to maintain high service availability, offering a more flexible and cost-efficient alternative to maintaining idle, hardware-based redundant systems (Compl. ¶25).

Key Claims at a Glance

• The complaint asserts independent Claim 6 (Compl. ¶100).
• The essential elements of Claim 6 include:
  • monitoring a first availability of a first service, the first service having a first availability requirement and a first availability tolerance;
  • detecting a reduction in the first availability of the first service;
  • creating capacity for the first service by deactivating a second service on a first active virtual machine on a server, the second service having a second availability exceeding a second availability tolerance and having a second availability requirement lower than the first availability requirement; and
  • activating a second active virtual machine executing the first service on the server. (’472 Patent, col. 9:32-45).
• The complaint also asserts dependent Claims 7-10 (Compl. ¶100).

U.S. Patent No. 7,313,102 - “System and Method for Subnet Configuration and Selection,” issued December 25, 2007

The Invention Explained

• Problem Addressed: The patent's background explains that prior art tools for managing IP address spaces were separate from provisioning systems and required network managers to manually determine appropriate IP addresses and network masks for network consumers. These tools did not manage inter-related characteristics like performance and security, which hindered resource allocation and increased error rates (’102 Patent, col. 1:29-48).
• The Patented Solution: The invention describes a method where subnets are first organized into groups based on "logical properties" such as security or performance (’102 Patent, col. 3:35-47). Network consumers are assigned access to specific subnet groups. The invention then provides for a "constrained selection" of a subnet via a graphical user interface (GUI) with selectable fields, which guides the consumer through choosing parameters like public/private address space, a gateway device, a subnet group, and a subnet mask. This process intelligently limits the available choices to prevent errors and streamline provisioning (’102 Patent, Abstract; col. 4:3-32).
• Technical Importance: The invention provided a more automated and error-resistant method for subnet provisioning by integrating policy (via logical groups) and user constraints (via the GUI) into a single workflow, reducing reliance on manual configuration (Compl. ¶38).

Key Claims at a Glance

• The complaint asserts independent Claim 1 (Compl. ¶151).
• The essential elements of Claim 1 include:
  • grouping the subnets into subnet groups based on logical properties of the subnets;
  • assigning to each network consumer those subnet groups that are accessible to that network consumer; and
  • providing for constrained selection of a particular subnet by a network consumer accomplished by way of a graphical user interface with selectable fields, wherein the constrained selection includes (i) selecting a public or private type address space, (ii) if applicable, selecting a gateway device..., (iii) selecting a subnet group..., and (iv) selecting a subnet mask that represents a size of the particular subnet. (’102 Patent, col. 9:14-32).
• The complaint also asserts dependent Claims 2-11 (Compl. ¶151).

U.S. Patent No. 7,649,912 - “Time Synchronization, Deterministic Data Delivery and Redundancy for Cascaded Nodes on Full Duplex Ethernet Networks,” issued January 19, 2010

Technology Synopsis

The patent addresses the difficulty of maintaining precise, sub-microsecond time synchronization across multiple "daisy-chain connected network nodes," a problem exacerbated by cumulative, random delays introduced by network switches in "store and forward" architectures (Compl. ¶¶45-47). The invention discloses a method where an intermediate node in the network receives a time synchronization message, adjusts the timestamp within the message to account for the internal switching delay, recalculates the message checksum and CRC code "on the fly," and forwards the corrected message, thereby improving clock synchronization performance (Compl. ¶¶48-49).

Asserted Claims

Independent Claims 1 and 7, and dependent Claims 2-3 and 5-6, 8-12 (Compl. ¶¶51, 193).

Accused Features

Broadcom’s ethernet switching products and associated software/firmware (e.g., StrataDNX devices, BroadPTP software, BroadSync firmware) that implement the IEEE 1588 Precision Time Protocol (PTP) for time synchronization (Compl. ¶193).

U.S. Patent No. 7,447,931 - “Step time change compensation in an industrial automation network,” issued November 4, 2008

Technology Synopsis

The patent addresses deficiencies in time synchronization protocols like IEEE 1588, which did not properly account for sudden "step changes" in a master clock's time (e.g., from a manual reset or loss of a reference signal), leading to timing errors across a network (Compl. ¶64). The invention provides a method for detecting such a step change by comparing successive time offsets received from a source and, if a change is detected, selectively updating a timestamp and associated offset to compensate for the change, thereby maintaining accurate timekeeping (Compl. ¶¶72-73).

Asserted Claims

Independent Claim 27 and dependent Claims 28-32 (Compl. ¶241).

Accused Features

Broadcom’s ethernet switching products (e.g., BCM56070, StrataDNX) and associated software that implement IEEE 1588-based time synchronization (Compl. ¶241).

U.S. Patent No. 7,656,751 - “Step time change compensation in an industrial automation network,” issued February 2, 2010

Technology Synopsis

Sharing a specification with the ’931 Patent, this patent addresses the same problem of compensating for step changes in a master clock's time (Compl. ¶78). The invention is claimed as a system comprising a "timestamp component" that captures timestamps and offsets from a network node, and a "time synch component" that identifies step changes to a master clock and synchronizes the local clock time of the network node with the identified step change (Compl. ¶86).

Asserted Claims

Independent Claim 1 and dependent Claims 2-14 (Compl. ¶306).

Accused Features

Broadcom’s ethernet switching products (e.g., BCM56070, StrataDNX) and associated software that implement IEEE 1588-based time synchronization (Compl. ¶306).

III. The Accused Instrumentality

Product Identification

• ’472 Patent: The "Broadcom Load Balancing Accused Products," which include cloud service platforms like VMware Cloud Foundation and products incorporating VMware NSX/NSX-T Data Center and VMware Avi Load Balancer (Compl. ¶100).
• ’102 Patent: The "Broadcom Subnet Provisioning Accused Products," which primarily consist of products incorporating VMware NSX/NSX-T Data Center (Compl. ¶151).

Functionality and Market Context

• The accused products are software-defined networking and virtualization platforms that allow customers to create, manage, and scale virtualized network and computing resources. The functionality accused of infringing the ’472 Patent relates to automatic scaling and migration of "virtual services" running on "Service Engines" (virtual machines) to manage network load and ensure service availability (Compl. ¶¶103-104, 107). The functionality accused of infringing the ’102 Patent relates to the tools within VMware NSX for creating and managing Virtual Private Clouds (VPCs) and provisioning network subnets through a graphical user interface (Compl. ¶¶154, 159).
• The complaint alleges these products are highly significant to Defendants' business, noting that VMware Cloud Foundation accounted for $2.5 billion in revenue in a single fiscal quarter (Compl. ¶119).

IV. Analysis of Infringement Allegations

’472 Patent Infringement Allegations

Claim Element (from Independent Claim 6)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
monitoring a first availability of a first service, the first service having a first availability requirement and a first availability tolerance;	The accused products display monitored "virtual services," which include a numeric, color-coded "health" status indicator.	¶103	col. 9:32-35
detecting a reduction in the first availability of the first service;	The products use metric-based thresholds, such as detecting when a Service Engine's CPU usage exceeds an 80% average, to determine that availability is reduced and trigger a scaling or migration operation.	¶¶105, 108	col. 9:36-37
creating capacity for the first service by deactivating a second service on a first active virtual machine on a server...	When CPU usage is high but no single service is dominant, the system performs a "migration" where the virtual service is moved to a new Service Engine, and the old Service Engine is subsequently "removed from the virtual service configuration."	¶¶109-110	col. 9:38-44
activating a second active virtual machine executing the first service on the server.	When a virtual service's traffic exceeds the capacity of a single Service Engine, the system performs a "scale out" operation by adding "one or more new SEs to the virtual service," which may involve "instantiating a new SE VM."	¶¶112-113	col. 9:44-45

Identified Points of Contention:

• Technical & Scope Questions: The complaint's theory for the "creating capacity" element relies on a "migration" operation where an instance of the first service is moved off an overloaded virtual machine, and that original virtual machine is later removed. This raises the question of whether this process meets the claim limitation of "deactivating a second service" to create capacity. The infringement analysis may focus on whether moving the service-at-issue is functionally and definitionally equivalent to deactivating a different, lower-priority service to make room, as the patent language suggests. The complaint's use of a flowchart depicting "Migrate any VS from SE" illustrates this alleged functionality (Compl. Fig. 8, ¶112).
• Technical Questions: What evidence does the complaint provide that the "migration process" creates capacity for the first service, as opposed to simply moving the first service to a different location with more capacity? The claim requires a causal link between deactivating the second service and creating capacity for the first, which will be a point of factual dispute.

’102 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
grouping the subnets into subnet groups based on logical properties of the subnets;	The VMware NSX products support and group subnets into "tier-0 and tier-1 subnets," which are described as groups based on logical properties.	¶154	col. 9:15-17
assigning to each network consumer those subnet groups that are accessible to that network consumer;	Users are assigned to an "NSX Virtual Private Cloud (VPC)" and can only "add subnets (networks) inside the NSX VPC that is assigned to them," thereby restricting access to certain subnet groups.	¶¶154, 157	col. 9:18-20
providing for constrained selection of a particular subnet by a network consumer accomplished by way of a graphical user interface with selectable fields, wherein the constrained selection includes...	A GUI is provided for adding a subnet, which allows a user to specify properties from limited options.	¶¶158-160	col. 9:21-32
(i) selecting a public or private type address space...	The user interface allows selection of an "Access Mode" of "Private, Public, [or] Isolated."	¶160	col. 9:24-25
(ii) if applicable, selecting a gateway device...	The user interface provides for "Gateway IP specification as part of subnet provisioning."	¶160	col. 9:26-28
(iii) selecting a subnet group...	The user selects from subnet groups (e.g., tier-0 or tier-1) made accessible within their assigned VPC.	¶154	col. 9:29-30
(iv) selecting a subnet mask that represents a size of the particular subnet.	The user interface allows selection of a subnet "size" from a drop-down menu and specification of an "IP CIDR," which represents the subnet mask.	¶¶159-160	col. 9:31-32

Identified Points of Contention:

• Scope Questions: Does the accused products' use of "tier-0" and "tier-1" groupings, which relate to routing architecture, constitute "grouping... based on logical properties" as contemplated by the patent, which provides examples such as "security characteristics and performance characteristics" (’102 Patent, col. 3:39-42)? The complaint's screenshot of an annotated administrator guide illustrates these "tier-0" groupings (Compl. Fig. 17, ¶154).
• Technical Questions: Does the workflow in the accused GUI for adding a subnet, as depicted in the complaint's screenshot of subnet properties (Compl. Fig. 20, ¶160), require the user to perform the specific four-part "constrained selection" in the sequence and manner claimed, or are there material differences in the technical operation?

V. Key Claim Terms for Construction

For the ’472 Patent:

• The Term: "deactivating a second service"
• Context and Importance: This term is critical to the infringement analysis for Claim 6. The Defendants may argue that the accused "migration" feature moves the first service itself and does not involve a distinct "second service," thereby avoiding infringement. The construction of this term will determine whether moving a single service off a VM to free up resources on that VM can satisfy the claim element.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification's background describes a general principle of diverting resources "from lower priority services to higher priority services to improve service availability" (’472 Patent, col. 1:65-2:3). This language could support an interpretation where any resource reallocation that deactivates a function to free up server capacity meets the spirit of the invention.
  • Evidence for a Narrower Interpretation: Claim 6 explicitly distinguishes between a "first service" and a "second service." The specification further states, "an instance of a lower priority service is deactivated to provide an available server. The higher priority service is then activated on the available server" (’472 Patent, col. 2:1-5). This language suggests two distinct services are involved in the claimed process.

For the ’102 Patent:

• The Term: "constrained selection"
• Context and Importance: The patentability of Claim 1, which the court has already found directed to ineligible subject matter, may hinge on whether "constrained selection" is interpreted as a specific, concrete technical process rather than an abstract idea of guiding a user. For infringement, the dispute will be whether the accused GUI implements this specific process.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The abstract describes providing "an interface for constrained selection of a particular subnet," and the specification states the mechanism "may comprise a graphical user interface with selectable fields" (’102 Patent, col. 3:59-61), which could support a general interpretation covering any guided GUI.
  • Evidence for a Narrower Interpretation: Claim 1 itself defines "constrained selection" by explicitly listing four included selection steps (address type, gateway, group, mask). The complaint notes that during prosecution, the examiner highlighted the novelty of including "(iv) selecting a subnet mask" as part of the constrained selection (Compl. ¶36), suggesting the entire combination is integral to the term's meaning and patentable scope.

VI. Other Allegations

• Indirect Infringement: For both the ’472 and ’102 patents, the complaint alleges inducement based on Defendants’ creation and distribution of marketing materials, installation guides, and interactive "VMware Hands-on Labs" which allegedly instruct and encourage customers to use the accused products in an infringing manner (Compl. ¶¶126-129, 172-175). Contributory infringement is alleged on the basis that the accused software products are material components specially made and adapted for practicing the claimed methods and have no substantial non-infringing use, particularly when automated features are enabled (Compl. ¶¶137-138, 182-183).
• Willful Infringement: Willfulness is alleged for all asserted patents. The claims are primarily based on Defendants' alleged pre-suit knowledge via notice letters sent by Plaintiff on December 23, 2024, and April 15, 2025 (Compl. ¶¶121-122). The complaint further alleges that VMware has a "culture of copying" and "willful blindness," citing adverse findings in prior, unrelated patent litigation as evidence (Compl. ¶¶146-148).

VII. Analyst’s Conclusion: Key Questions for the Case

• Patent Eligibility: As the complaint acknowledges that the court has already found asserted claims from all five patents directed to ineligible subject matter, the primary legal question is whether this ruling will be upheld on appeal. The case will test whether claims directed to automated management of virtual computing resources and network configurations are viewed as patent-ineligible abstract ideas or as concrete technical improvements to computer functionality.
• Functional Equivalence (’472 Patent): A central evidentiary question will be one of functional equivalence: does the accused product’s "migration" feature, which relocates an instance of a service, perform the specific function of "deactivating a second service" to create capacity for a "first service" as required by Claim 6? The analysis may turn on whether moving the same service is technically and legally equivalent to sacrificing a different service.
• Definitional Scope (’102 Patent): A core issue will be one of definitional scope: can the accused products' "tier-0" and "tier-1" router-based classifications be construed as the "grouping... based on logical properties" recited in Claim 1? Further, the case will likely examine whether the accused GUI's workflow for adding a subnet embodies the specific four-part "constrained selection" process required by the claim or if there is a fundamental mismatch in technical operation.