DCT
2:25-cv-01859
Starbucks Corp v. Valtrus Innovations Ltd
Key Events
Complaint
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Starbucks Corporation (Washington)
- Defendant: Valtrus Innovations Limited (Ireland) and Key Patent Innovations Limited (Ireland)
- Plaintiff’s Counsel: Kilpatrick Townsend & Stockton LLP; Baker Botts LLP
- Case Identification: 2:25-cv-01859, W.D. Wash., 09/25/2025
- Venue Allegations: Plaintiff alleges venue is proper in the Western District of Washington because Defendants purposefully directed a patent enforcement campaign into the state, including sending a demand letter and subsequent communications to Plaintiff's headquarters in Seattle.
- Core Dispute: Plaintiff seeks a declaratory judgment of non-infringement and invalidity for eleven patents asserted by Defendants, alleging that its use of common open-source software platforms does not practice the patented technologies. Plaintiff also asserts claims under Washington's Patent Troll Prevention Act for bad-faith patent assertion.
- Technical Context: The technologies at issue involve foundational open-source platforms for large-scale data processing, cloud computing, and distributed systems, which are critical components of modern digital services infrastructure.
- Key Procedural History: The complaint was filed following a seven-month period of licensing negotiations initiated by a demand letter from Defendants. Plaintiff alleges that Defendants are non-practicing entities, that Valtrus is an undercapitalized shell company for its parent KPI, and that the infringement assertions were made in bad faith without adequate pre-suit investigation, pointing to what it characterizes as boilerplate demands and blanket accusations against general-purpose software.
Case Timeline
| Date | Event |
|---|---|
| 2003-07-25 | U.S. Patent No. 7,936,738 Priority Date |
| 2006-04-26 | U.S. Patent No. 8,370,416 Priority Date |
| 2010-06-01 | U.S. Patent No. 7,730,218 Issued |
| 2011-03-08 | U.S. Patent No. 7,904,686 Issued |
| 2011-05-03 | U.S. Patent No. 7,936,738 Issued |
| 2013-02-05 | U.S. Patent No. 8,370,416 Issued |
| 2022-01-20 | U.S. Patent No. 6,889,244 Expired |
| 2023-07-17 | U.S. Patent No. 7,251,588 Expired |
| 2023-10-25 | U.S. Patent No. 7,376,953 Expired |
| 2025-02-04 | Valtrus sends Demand Letter to Starbucks |
| 2025-02-14 | Valtrus counsel follows up with Starbucks |
| 2025-05-22 | Valtrus counsel follows up with Starbucks |
| 2025-08-17 | Valtrus counsel follows up with Starbucks |
| 2025-09-17 | Phone call between Starbucks and Valtrus representatives |
| 2025-09-23 | Multiple emails exchanged between parties |
| 2025-09-24 | Phone conversation and PowerPoint presentation |
| 2025-09-25 | Valtrus sends Starbucks claim charts |
| 2025-09-25 | Complaint for Declaratory Judgment filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 7,936,738 - "Fault Tolerant Systems"
The Invention Explained
- Problem Addressed: The patent describes the challenge of maintaining "task preservation" in high-availability (HA) systems, such as telecommunications networks, after a fault or switchover from an active to a standby system. It notes that conventional solutions requiring a common storage element to store task context data add complexity and cost to such systems (ʼ738 Patent, col. 1:5-col. 2:66).
- The Patented Solution: The invention proposes a method to store this context information directly within an outgoing message sent between nodes via a protocol stack. By embedding the context in the message, a response to that message will contain the necessary information, which can then be used to restore the context of a protocol layer on a newly activated standby system, thereby eliminating the need for a separate, shared storage element (ʼ738 Patent, Abstract; col. 2:1-11; Fig. 5).
- Technical Importance: This approach aimed to simplify the design and reduce the cost of fault-tolerant systems by decentralizing state management and leveraging the message traffic itself for context preservation (ʼ738 Patent, col. 2:63-66).
Key Claims at a Glance
- The complaint asserts non-infringement of independent claim 1 (Compl. ¶164).
- Essential elements of claim 1 include:
- providing the outgoing message from an application to a layer of the protocol stack
- selectively indicating to the layer that context information is to be obtained for that layer
- obtaining context information in accordance with the indication
- adding the obtained context information to the outgoing message such that a response, received from the destination node, to the outgoing message contains the obtained context information
- The complaint does not explicitly reserve the right to assert dependent claims for this patent.
U.S. Patent No. 8,370,416 - "Compatibility Enforcement in Clustered Computing Systems"
The Invention Explained
- Problem Addressed: The patent addresses technical issues in clustered computing systems, specifically ensuring compatibility between hardware, software, and firmware, and ensuring compliance with licensing terms associated with individual devices and the cluster as a whole (ʼ416 Patent, col. 1:25-32).
- The Patented Solution: The invention discloses a method for creating a clustered system where "license information" is stored in a memory module on a computing node. This information includes a "bundle-type parameter" that identifies characteristics like the permissible size of the cluster. The system then activates the cluster only when the number of nodes complies with the rules defined by this bundle-type parameter, creating a technical mechanism for license enforcement (ʼ416 Patent, Abstract; col. 2:35-44).
- Technical Importance: The described technology provides a method for enforcing complex, bundle-based software licensing models directly within the architecture of a scalable, clustered computing environment (ʼ416 Patent, col. 5:45-54).
Key Claims at a Glance
- The complaint asserts non-infringement of independent claim 1 (Compl. ¶172).
- Essential elements of claim 1 include:
- storing license information for a computing cluster in a memory module associated with a computing node, wherein the licensing information includes a bundle-type parameter
- initializing the computing cluster in a first computing node
- adding one or more available computing nodes to the computing cluster
- activating the computing cluster when the computing cluster includes a number of nodes that complies with the bundle-type parameter
- The complaint does not explicitly reserve the right to assert dependent claims for this patent.
U.S. Patent No. 6,889,244 - "Method and Apparatus for Passing Messages Using a Fault Tolerant Storage System"
- Technology Synopsis: The patent describes a method for transmitting messages between nodes using a fault-tolerant storage system (FTSS). The FTSS is used not just for storage but also as a reliable transport for messages, which are stored in a data structure on the FTSS's highly reliable media (Compl. ¶91).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶87).
- Accused Features: "All offerings using Apache Kafka" (Compl. ¶87).
U.S. Patent No. 7,120,832 - "Storage Device Performance Monitor"
- Technology Synopsis: The patent discloses a method for monitoring the performance of a storage device by intercepting communications between a computer system and the device, analyzing those communications, and reallocating data on the device to enhance performance based on the analysis (Compl. ¶104).
- Asserted Claims: Independent claim 25 is asserted (Compl. ¶98).
- Accused Features: "All offerings using Apache HBase," "All offerings using Apache Cassandra," and "All offerings using RocksDB Universal Compaction" (Compl. ¶98).
U.S. Patent No. 7,251,588 - "System for Metric Introspection in Monitoring Sources"
- Technology Synopsis: The technology relates to a monitoring system where metric definitions for a monitored component are stored in a machine-readable format. A monitoring tool can then access these definitions to understand the data it is collecting and process it accordingly (Compl. ¶114).
- Asserted Claims: Independent claim 13 is asserted (Compl. ¶110).
- Accused Features: "All offerings using Kubernetes" (Compl. ¶110).
U.S. Patent No. 7,376,953 - "Apparatus and Method for Routing a Transaction to a Server"
- Technology Synopsis: The patent describes a method for routing a transaction to a front-end server by identifying an attribute-based category for the transaction and then selecting a server assigned to handle that category, with a fallback to a default server if none is identified (Compl. ¶125).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶121).
- Accused Features: "All offerings using Apache Kafka" (Compl. ¶121).
U.S. Patent No. 7,640,332 - "System and Method for Hot Deployment/Redeployment in Grid Computing Environment"
- Technology Synopsis: The patent discloses a method for hot deployment of applications in a grid computing environment. The method involves adding a new application version to a repository, using a discovery module to identify which grid nodes are running the application, and notifying a client application manager on those nodes to deploy the new version (Compl. ¶136).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶133).
- Accused Features: "All offerings using Kubernetes" (Compl. ¶133).
U.S. Patent No. 7,730,218 - "Method and System for Configuration and Management of Client Access to Network-Attached-Storage"
- Technology Synopsis: This patent describes a system where a "master-agent" executable constructs and forwards an executable code block to a target computer system. When executed, this code block connects the target system to a network-attached-storage (NAS) object (Compl. ¶147).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶144).
- Accused Features: "All offerings using Apache Hadoop" (Compl. ¶144).
U.S. Patent No. 7,904,686 - "Data Security for Use with a File System"
- Technology Synopsis: The patent discloses a data security method for a file system that applies a mapping function to data block numbers contained in an index node (inode). The resulting mapped numbers correspond to the actual storage addresses of the file's data, obscuring the direct link between the inode and physical storage (Compl. ¶155).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶152).
- Accused Features: "All offerings using Apache Hadoop" (Compl. ¶152).
U.S. Patent No. 8,379,538 - "Model-Driven Monitoring Architecture"
- Technology Synopsis: The patent describes a method using a machine-readable monitoring "model" that defines the configuration of a monitoring environment. When the environment's configuration changes, the model is updated, and an element of the environment reads the model to autonomously adapt its operation (Compl. ¶181).
- Asserted Claims: Independent claim 7 is asserted (Compl. ¶178).
- Accused Features: "All offerings using Kubernetes" (Compl. ¶178).
U.S. Patent No. 8,965,843 - "Prioritized Replication Paths"
- Technology Synopsis: The technology relates to updating data across multiple nodes. The method involves determining a node for an update based on node priority and determining the version of data to send by comparing the target node's data version with that of another node having a higher priority (Compl. ¶191).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶188).
- Accused Features: "All offerings using Apache Cassandra" (Compl. ¶188).
III. The Accused Instrumentality
Product Identification
- The complaint does not identify a specific Starbucks product, service, or method. Instead, it addresses the allegation made in Defendants' demand letter that "all offerings" by Starbucks that utilize certain open-source software platforms infringe the patents-in-suit (Compl. ¶¶ 33-34, 87, 171). These platforms include Apache Kafka, Kubernetes, Apache Hadoop, Apache Cassandra, RocksDB, and Apache HBase (Compl. ¶20).
Functionality and Market Context
- The complaint describes the accused open-source platforms as "staple, general-purpose" components for modern computing infrastructure (Compl. ¶49). It alleges that Starbucks uses these platforms in conjunction with cloud-based hosting services such as Amazon Web Services and Microsoft Azure (Compl. ¶20). The functionality of each platform is described at a high level; for example, Apache Kafka is identified as an "open-source event streaming platform" (Compl. ¶89), and Kubernetes is an "open-source system for automating deployment, scaling, and management of containerized applications" (Compl. ¶59). The complaint's central premise is that Defendants' allegations target Starbucks's use of these standard, widely-adopted technologies rather than any specific, proprietary implementation by Starbucks (Compl. ¶¶ 55, 81).
No probative visual evidence provided in complaint.
IV. Analysis of Infringement Allegations
[7,936,738] Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| providing, by the computing device, the outgoing message from an application to a layer of the protocol stack... | The complaint alleges that Apache Kafka's functionality does not meet this limitation because its use of user-level headers is an application-level construct, not an interaction with a "layer of the protocol stack" as claimed. | ¶165 | col. 8:3-7 |
| selectively indicating to the layer of the protocol stack that context information is to be obtained for that layer; | The complaint asserts that Kafka provides no mechanism for an application to instruct a network or protocol stack layer to collect layer-specific context. | ¶167 | col. 8:8-11 |
| adding... the obtained context information to the outgoing message such that a response... contains the obtained context information. | The complaint argues that Apache Kafka does not guarantee that responses will echo the application-supplied context headers, as required to ensure the context is returned. | ¶166 | col. 8:14-19 |
Identified Points of Contention:
- Scope Questions: A principal question will be whether the term "layer of the protocol stack," described in the patent in the context of fault-tolerant telecommunication systems, can be construed to read on the application-level header mechanism in a distributed messaging platform like Apache Kafka.
- Technical Questions: The complaint raises the question of whether Kafka's standard operation performs the claimed functions. Specifically, what evidence supports the allegation that Kafka's user-level headers constitute a "selective indication to the layer" to obtain "context information," and what mechanism ensures that a response message "contains" that same information for restoration purposes?
[8,370,416] Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| storing license information for a computing cluster in a memory module associated with a computing node, wherein the licensing information includes a bundle-type parameter... | The complaint alleges that Kubernetes, as an open-source system, does not store "license information" or use "bundle-type parameters" as contemplated by the patent. | ¶¶61, 62, 174 | col. 10:55-61 |
| initializing the computing cluster in a first computing node; | The complaint does not contest that Kubernetes initializes clusters. | - | col. 10:62-63 |
| adding one or more available computing nodes to the computing cluster; | The complaint does not contest that nodes can be added to a Kubernetes cluster. | - | col. 10:64-65 |
| activating the computing cluster when the computing cluster includes a number of nodes that complies with the bundle-type parameter. | The complaint alleges that standard Kubernetes control planes do not enforce cluster activation based on a node-count license or any "bundle-type parameter," as Kubernetes is free and does not use license keys to gate its operation. | ¶¶65, 66, 175 | col. 10:65-col. 11:2 |
Identified Points of Contention:
- Scope Questions: A key issue will be the definition of "license information." Does this term require a commercial licensing construct as described in the patent's specification, or could it be interpreted more broadly to cover any configuration parameter within Kubernetes that might affect cluster size or composition?
- Technical Questions: What evidence does the infringement theory provide that any component or configuration in a standard Kubernetes deployment performs the function of "activating the computing cluster" only when a node count "complies with the bundle-type parameter"? The complaint alleges a fundamental mismatch between the claimed license-gated activation flow and the actual operation of Kubernetes (Compl. ¶¶ 62, 68).
V. Key Claim Terms for Construction
The Term: "a layer of the protocol stack" (from ’738 Patent, Claim 1)
Context and Importance: This term is critical because Starbucks's non-infringement argument hinges on its assertion that Apache Kafka operates at the application level, whereas the patent is directed to operations at a lower-level "protocol stack" layer (Compl. ¶165). The construction of this term may determine whether the patent's claims can reach application-level software architectures.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claims themselves do not specify a particular type of protocol stack (e.g., OSI, TCP/IP). A party might argue that any system with a hierarchical structure of discrete layers for communication, including application-level protocols, could be considered a "protocol stack" in a general sense.
- Evidence for a Narrower Interpretation: The specification describes the invention in the context of high-availability systems like telephony networks (ʼ738 Patent, col. 1:47-53). Figure 4 of the patent depicts distinct "APPLICATION," "SIGNALING," and "TRANSPORT" layers, suggesting a more formal, traditional networking or telecommunications stack rather than general application logic.
The Term: "license information" (from ’416 Patent, Claim 1)
Context and Importance: The existence of "license information" is a predicate for infringement. Starbucks alleges that Kubernetes, being open-source, does not use "license keys" or any equivalent "license information" to control cluster activation (Compl. ¶¶61, 66). The definition of this term will be central to determining if any configuration data within Kubernetes can satisfy this claim limitation.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The term is not explicitly defined. An argument could be made that any data that enforces a rule or constraint on the system—such as a configuration file setting a maximum node count—could functionally be considered "license information," regardless of a commercial context.
- Evidence for a Narrower Interpretation: The specification consistently frames "license information" in a commercial context, referring to "bundle-types" as related to sales, "license keys," and an "external license server" (ʼ416 Patent, col. 5:45-67; col. 7:1-6). The abstract specifies the information includes a "bundle-type parameter that identifies a characteristic of the cluster," suggesting a commercial package or SKU rather than a simple technical limit.
VI. Other Allegations
- Indirect Infringement: The complaint anticipates and counters potential indirect infringement theories, arguing that Defendants' demand letter fails to plead the required elements of knowledge and intent (Compl. ¶53). It further raises the defense of divided infringement, noting that core steps of the accused open-source platforms are often performed by third-party cloud providers, not Starbucks, which would require Defendants to prove direction or control under the Akamai standard (Compl. ¶¶56, 90, 117).
- Willful Infringement: As a declaratory judgment action filed by the accused infringer, willfulness is not a primary claim. However, the complaint's allegations of "bad-faith patent assertion" under Washington's PTPA serve a similar function. The basis for these allegations includes the non-specific nature of the infringement claims targeting generic open-source platforms, the use of allegedly boilerplate demand letters sent to disparate companies, misrepresentation of patent ownership, and the assertion of claims against technologies that allegedly have a "basic mismatch" with the patented invention (Compl. ¶¶49, 66, 72).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of definitional scope: can claim terms rooted in specific technical contexts, such as "a layer of the protocol stack" from telecommunications systems or "license information" from commercial software clustering, be construed broadly enough to cover the distinct functionalities of general-purpose, open-source platforms like Apache Kafka and Kubernetes?
- A key question for the statutory claims will be the sufficiency of pre-suit diligence: does the defendants' alleged strategy of asserting infringement against "all offerings using" widely adopted open-source software, without identifying specific accused products or providing detailed infringement mappings, meet the good-faith requirements of patent law, or does it constitute a basis for liability under Washington's Patent Troll Prevention Act?
- An underlying evidentiary question will be one of functional and architectural mismatch: beyond claim construction, the case will likely turn on factual determinations of whether the accused open-source platforms, as used by the plaintiff, actually perform the specific steps and contain the specific components required by the asserted method claims, or if there is a fundamental difference in their technical operation and architecture.