Intellectual Ventures I LLC v. Nationwide Mutual Insurance Co

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Intellectual Ventures I LLC and Intellectual Ventures II LLC (Delaware)
  • Defendant: Nationwide Mutual Insurance Company (Ohio)
  • Plaintiff's Counsel: Cherry Johnson Siegmund James PC; Kasowitz LLP
• Case Identification: 3:25-cv-00632, N.D. Tex., 03/02/2026
• Venue Allegations: Plaintiff alleges venue is proper in the Northern District of Texas because Defendant maintains numerous regular and established places of business in the district, including agency offices and the residences of remote employees, and has committed acts of infringement there.
• Core Dispute: Plaintiff alleges that Defendant's backend computing systems, which utilize technologies such as Docker and Apache Spark to support its insurance and financial services, infringe three patents related to distributed computing, parallel processing, and application environment management.
• Technical Context: The patents relate to foundational technologies for managing and executing applications in large-scale, multi-computer environments, a critical component of modern enterprise cloud and data processing infrastructure.
• Key Procedural History: The complaint alleges Defendant had pre-suit knowledge of all three patents-in-suit via notice letters. A letter dated July 1, 2024, allegedly provided notice of the '844 Patent, and a letter dated March 14, 2025, allegedly provided notice of the '282 and '080 Patents.

Case Timeline

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

U.S. Patent No. 7,721,282 - "Block-Level I/O Subsystem For Distributed Application Environment Management" (Issued May 18, 2010)

The Invention Explained

• Problem Addressed: In clustered computing environments, updating the boot or application images for all computers ("compute nodes") is cumbersome and inefficient, as it requires updating numerous full copies of the image (Compl. ¶47, citing '282 Patent, col. 1:54-57).
• The Patented Solution: The invention proposes a system where a single, read-only "root image" (containing the base operating system and applications) is stored centrally Compl. ¶47 Each compute node uses a separate, writable "leaf image" that stores only the unique changes and new data blocks specific to that node '282 Patent, abstract A "union block device" acts as an intermediary, merging the blocks from the central root image and the specific leaf image on-the-fly to create a complete, functional application environment for the compute node without creating a full copy '282 Patent, col. 2:24-41
• Technical Importance: This approach significantly reduces storage requirements and simplifies updates in large-scale distributed systems by centralizing the common data and only distributing the differences.

Key Claims at a Glance

• The complaint asserts at least independent claim 1 Compl. ¶51
• Essential Elements of Claim 1:
  • A compute node comprising a computer system.
  • A first storage unit for storing blocks of a root image.
  • A second storage unit for storing a leaf image containing new data and changes to the root image.
  • A union block device that interfaces between the compute node and the storage units.
  • The union block device creates the application environment by merging the blocks of the root image with the blocks of the leaf image.
  • The union block device comprises a low-level driver for interfacing between the storage units and the file system.
  • The union block device, upon receiving a write request, creates a persistent mapping for the data sector being written.
• The complaint does not explicitly reserve the right to assert dependent claims.

U.S. Patent No. 7,712,080 - "Systems and Methods for Parallel Distributed Programming" (Issued May 4, 2010)

The Invention Explained

• Problem Addressed: The patent identifies challenges in programming for parallel distributed systems. Existing methods were either too complex and altered the original algorithm's structure ("message passing") or were easier to program but less efficient due to large data transfers ("distributed shared memory") (Compl. ¶88, citing '080 Patent, col. 1:31-col. 2:2).
• The Patented Solution: The invention describes a system using a "distributed shared variable"-a single logical variable whose data is physically spread across multiple memories-and a "distributed sequential computing program" '080 Patent, abstract This program is configured to transform into a "distributed parallel computing program" by "spawning" a child program when a specific intermediate condition is met. The original (parent) program and the newly spawned child program then execute concurrently to perform parallel processing '080 Patent, col. 11:15-29
• Technical Importance: This technology aims to provide a higher-level programming model that simplifies the development of efficient parallel applications by automating the transformation from sequential to parallel execution based on data needs and program state.

Key Claims at a Glance

• The complaint asserts at least independent claim 9 Compl. ¶92
• Essential Elements of Claim 9:
  • A distributed parallel computing system with at least one memory area and at least one processor.
  • At least one distributed shared variable capable of being loaded into the memory area.
  • At least one distributed sequential computing program configured to operate on the processor and access the shared variable.
  • The program is configured to transform into a distributed parallel program by spawning at least one child program when an intermediate condition occurs.
  • The parallel program concurrently uses the original program and the spawned child program to perform parallel processing.
  • The intermediate condition comprises a result that is required by the spawned child program to continue its computation.
• The complaint does not explicitly reserve the right to assert dependent claims.

U.S. Patent No. 8,332,844 - "Root Image Caching and Indexing for Block-level Distributed Application Management" (Issued December 11, 2012)

• Technology Synopsis: This patent builds upon the technology of the '282 Patent, addressing the same problem of inefficient boot image management in computer clusters (Compl. ¶124, citing '844 Patent, col. 1:64-67). The patented solution adds two key optimizations: caching blocks of the root image that have been frequently accessed to speed up delivery to other nodes, and a method for one compute node to index the root image and share those indexing results with other nodes to avoid redundant processing '844 Patent, abstract
• Asserted Claims: The complaint asserts at least independent claim 7 Compl. ¶128
• Accused Features: Nationwide's systems and services that utilize Docker containerization technology Compl. ¶125

III. The Accused Instrumentality

Product Identification

• The complaint identifies "Accused Systems and Services" used by Nationwide, which are alleged to be backend and customer-facing financial and insurance technologies that utilize open-source platforms, primarily Docker and Apache Spark Compl. ¶9 Compl. ¶39

Functionality and Market Context

• The accused systems allegedly support over 850 of Nationwide's business applications, including its claims and policy systems and its main website, Nationwide.com Compl. ¶50 Compl. ¶91
• The complaint alleges the use of Docker for containerization, which packages applications into isolated environments Compl. ¶54 The infringement theory focuses on Docker's use of a layered filesystem (OverlayFS), where a read-only base image is combined with a writable "container layer" that stores all changes Compl. ¶¶60-65 A diagram in the complaint illustrates the Docker architecture, including the Docker Host which manages images and containers Compl. p. 25
• The complaint alleges the use of Apache Spark for large-scale data processing Compl. ¶¶90-91 The infringement theory centers on Spark's architecture, where a "driver program" creates a plan to execute "various parallel operations on a cluster" by partitioning data into Resilient Distributed Datasets (RDDs) and launching parallel tasks across different nodes Compl. ¶95 Compl. ¶100 The complaint includes a diagram depicting the "Spark - Execution Workflow," which shows a driver program initiating transformations that result in parallel tasks Compl. p. 53

IV. Analysis of Infringement Allegations

'282 Patent Infringement Allegations

• Identified Points of Contention:
  • Scope Questions: A central dispute may arise over the term "union block device". The complaint maps this term to Docker's "overlay2" storage driver, which operates at the file-system level. The defense may argue that the patent, titled a "Block-Level I/O Subsystem" and claiming a "low-level driver," requires a device operating at the block level, below the file system, creating a potential mismatch with Docker's architecture.

'080 Patent Infringement Allegations

• Identified Points of Contention:
  • Technical Questions: The infringement theory equates Spark's scheduling of tasks via a Directed Acyclic Graph (DAG) with the patent's language of a program "transforming... by spawning at least one child." A key technical question will be whether this model of a central scheduler dispatching tasks is functionally the same as the parent-child spawning model described in the patent. The definition of the "intermediate condition" that triggers the transformation will be a critical point of analysis.

V. Key Claim Terms for Construction

Term 1 ('282 Patent)

• The Term: "union block device"
• Context and Importance: This term is the lynchpin of the infringement allegations against Docker for both the '282 and '844 patents. The case may turn on whether this term can be construed broadly enough to cover a file-system-level technology like Docker's OverlayFS driver, or if it is limited to a true block-level device driver.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The claim describes the device's function as "merging the blocks" from two sources to "create the application environment." Plaintiff may argue that any component performing this block-merging function, regardless of its position in the I/O stack, meets the claim's functional definition.
  • Evidence for a Narrower Interpretation: The patent's title is "Block-Level I/O Subsystem," and claim 1 explicitly calls it a "low-level driver for interfacing between the first and second storage units and the file system." This language suggests the device operates below the file system, not as part of it, potentially limiting the claim's scope to kernel-level block device drivers.

Term 2 ('080 Patent)

• The Term: "transforming... by spawning at least one child distributed sequential computing program"
• Context and Importance: This phrase is central to mapping the functionality of Apache Spark to the claims of the '080 Patent. The dispute will likely focus on whether Spark's model of a driver submitting a graph of tasks to a cluster manager/scheduler constitutes "spawning a child."
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The patent describes a transformation from a sequential program to a parallel one that performs computations concurrently '080 Patent, col. 5:44-47 Plaintiff may argue that this is a functional description of what the Spark driver and its associated tasks achieve.
  • Evidence for a Narrower Interpretation: The specification's figures and discussion of a "clone_hop()" command suggest a model where a process directly clones itself to migrate and continue work, akin to a "fork()" system call '080 Patent, Fig. 3b Defendant may argue this is technically distinct from Spark's more complex, scheduler-based execution model.

VI. Other Allegations

• Indirect Infringement: The complaint alleges inducement of infringement, stating that Nationwide provides instructions to its employees, partners, and vendors on how to use the accused systems that incorporate Docker and Spark Compl. ¶78 Compl. ¶116 Compl. ¶162 It also alleges contributory infringement, asserting the accused systems are especially adapted for infringement and not staple articles of commerce Compl. ¶82 Compl. ¶118 Compl. ¶166
• Willful Infringement: Willfulness is alleged for all three patents. The claims are based on alleged pre-suit actual knowledge from specific notice letters: one dated July 1, 2024 for the '844 Patent, and another dated March 14, 2025 for the '282 and '080 Patents Compl. ¶76 Compl. ¶112 Compl. ¶155 The complaint alleges that Nationwide's continued use of the accused technology after receiving these notices constitutes willful, deliberate, and flagrant infringement Compl. ¶158

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

• A primary technical-legal question will be one of definitional scope: can the term "union block device," described in the '282 Patent as a "low-level driver" operating below the file system, be construed to encompass a file-system-level union mount technology like Docker's OverlayFS storage driver? The outcome of this construction will be pivotal for the infringement claims related to both the '282 and '844 patents.
• Another core issue will be one of functional equivalence: does the Apache Spark execution model, which involves a driver program creating a directed acyclic graph (DAG) of tasks for a cluster scheduler, meet the specific '080 Patent claim requirement of a sequential program "transforming... by spawning at least one child"? The court's analysis of this technical comparison will determine the viability of the infringement claim against Spark.
• Given the complaint's specific allegations of pre-suit notice letters with exact dates for all three patents, a key factual question will be the nature and sufficiency of Nationwide's pre-suit knowledge. The case will examine whether Nationwide's continued use of the accused technologies after receiving notice was objectively reckless, which is central to the claim for willful infringement and potential enhanced damages.