SAP Se v. Trayport Ltd

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: SAP Se (Germany); Business Objects Software Ltd. t/a SAP Solutions (Ireland); Sybase, Inc. (Delaware)
  • Defendant: Trayport Limited (England and Wales)
  • Plaintiff’s Counsel: Ropes & Gray LLP
• Case Identification: 1:25-cv-02075, N.D. Ill., 02/27/2025
• Venue Allegations: Venue is alleged to be proper because the defendant is a foreign corporation, for which venue exists in any judicial district. The complaint also alleges specific contacts with the Northern District of Illinois, including sales to a Chicago-based customer (IncubEx), targeting of Illinois-based customers, and promotion of the accused products at industry conferences in Chicago.
• Core Dispute: Plaintiffs allege that Defendant's data and trading platform for wholesale energy markets infringes three patents related to real-time data stream processing, virtual database generation from heterogeneous sources, and data connection management.
• Technical Context: The technology at issue addresses the challenge of processing and unifying vast, continuous streams of data from disparate sources, a critical function in high-speed environments like financial and energy trading markets.
• Key Procedural History: The complaint does not mention any prior litigation, inter partes review proceedings, or licensing history related to the Asserted Patents.

Case Timeline

Date	Event
2005-02-03	U.S. Patent No. 8,396,886 Priority Date
2005-07-01	U.S. Patent No. 8,577,927 Priority Date
2011-12-20	U.S. Patent No. 9,009,354 Priority Date
2013-03-12	U.S. Patent No. 8396886 Issue Date
2013-11-05	U.S. Patent No. 8577927 Issue Date
2015-04-14	U.S. Patent No. 9009354 Issue Date
2022 (approx.)	Trayport partners with IncubEx to launch The Voluntary Climate Marketplace
2025-02-27	Complaint Filing Date

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

U.S. Patent No. 8,396,886 - "Continuous processing language for real-time data streams"

• Patent Identification: U.S. Patent No. 8,396,886, "Continuous processing language for real-time data streams," issued March 12, 2013. (Compl. ¶15).

The Invention Explained

• Problem Addressed: The patent’s background section describes conventional relational databases as being poorly suited for Complex Event Processing (CEP) applications, such as program trading, which require analysis of high-volume, real-time data streams. Traditional databases are optimized for querying static, stored data and become a bottleneck, while prior continuous query languages were overly complex and difficult to scale. (’886 Patent, col. 1:22-2:65).
• The Patented Solution: The invention is a specialized, SQL-like computer language for expressing queries that operate continuously and directly on data streams. It is based on a "publish/subscribe" model where queries subscribe to input data streams and publish results to output streams, avoiding the need for intermediate storage in a traditional database. The language includes clauses for defining time-based or row-based "windows" of data within a stream, allowing for operations on discrete segments of the continuous flow. (’886 Patent, Abstract; col. 3:30-56).
• Technical Importance: This approach aimed to simplify the development of scalable, high-performance CEP applications by providing developers with a familiar query language structure tailored to the unique demands of real-time stream processing. (’886 Patent, col. 2:51-56).

Key Claims at a Glance

• The complaint asserts independent claim 1. (Compl. ¶25).
• Essential elements of claim 1 include:
  • A computer-implemented method in a continuous processing system, comprising:
  • subscribing to one or more continuous input data streams from publishers, with each stream published in a publisher-specific format and protocol;
  • executing a query operation directly on the input data streams, without any pre-processing and without storing the streams in any data structure prior to execution, on a continuous basis; and
  • publishing a result in real time to output data streams in a subscriber-specific format and protocol, where the publisher and subscriber protocols vary.
• The complaint asserts "at least claim 1" and the prayer for relief seeks judgment on "one or more claims," which may suggest a reservation of rights to assert additional claims. (Compl. ¶25; Prayer for Relief ¶A).

U.S. Patent No. 8,577,927 - "Producing a virtual database from data sources exhibiting heterogeneous schemas"

• Patent Identification: U.S. Patent No. 8,577,927, "Producing a virtual database from data sources exhibiting heterogeneous schemas," issued November 5, 2013. (Compl. ¶16).

The Invention Explained

• Problem Addressed: The patent addresses the significant technical challenge of integrating data from multiple, disparate sources that possess different data structures and schemas (i.e., are "heterogeneous"). Manually programming a system to access such sources as a single, unified virtual database is described as a complex, delicate, and time-consuming task. (’927 Patent, col. 2:3-33).
• The Patented Solution: The invention provides a method to create a virtual database by first defining an "individual publication schema" for each heterogeneous source, which maps its native data into a "non-permanent structured instance of data." An "integration schema" is then established, which contains a set of rules for combining the data from these instances, analyzing and resolving conflicts, and defining a final target view for the user. This process is illustrated in the patent's figures, such as Figure 3, which depicts the workflow from source registration to integration view generation. (’927 Patent, Abstract; Fig. 3).
• Technical Importance: The method provides a structured, semi-automated approach to data integration, allowing users to query multiple, diverse data systems as if they were a single, coherent database without physically consolidating the underlying data. (’927 Patent, col. 2:36-44).

Key Claims at a Glance

• The complaint asserts independent claim 19. (Compl. ¶39).
• Essential elements of claim 19 include:
  • A method for implementation by one or more data processors comprising:
  • defining an individual publication schema for each of a plurality of heterogeneous sources, specifying mapping rules to a non-permanent structured instance of data;
  • generating said non-permanent structured instance of data;
  • establishing an integration schema with rules for combining data from each non-permanent instance, where the schema is used to analyze conflicts and defines a source view that includes hidden attributes; and
  • forming a target non-permanent structured instance of data from the heterogeneous sources based on the integration schema.
• The complaint’s phrasing suggests a reservation of rights to assert additional claims. (Compl. ¶39; Prayer for Relief ¶A).

U.S. Patent No. 9,009,354 - "Services and management layer for diverse data connections"

• Patent Identification: U.S. Patent No. 9,009,354, "Services and management layer for diverse data connections," issued April 14, 2015. (Compl. ¶17).
• Technology Synopsis: This patent addresses the complexity of application development on platforms that must support diverse data connection types (e.g., HTTP, RFC, Web Services). The invention provides a "common management layer" that abstracts the specifics of each connection type, offering common services like error handling and monitoring. This layer provides a "common entry point" for an application to send data, relieving the application's own logic from needing to handle the unique requirements of each connection protocol. (’354 Patent, Abstract; col. 4:15-20).
• Asserted Claims: The complaint asserts independent claim 11. (Compl. ¶54).
• Accused Features: The complaint alleges that Trayport's platform infringes by providing a management layer for diverse data connections and a common entry point to receive data destined for external receivers, thereby making the connections independent of the software application's logic. (Compl. ¶54).

III. The Accused Instrumentality

Product Identification

The complaint names a suite of "Accused Instrumentalities," including Joule, Autotrader, Data Analytics, Automated Trading, and others that constitute a "network and data platform for wholesale energy markets." (Compl. ¶¶18-19). The allegations appear to focus primarily on the "Joule" trading platform. (Compl. ¶20).

Functionality and Market Context

The Accused Instrumentalities are alleged to enable traders, brokers, and exchanges to engage in trading across multiple asset classes and markets. (Compl. ¶18). The Joule platform is specifically described as offering an "enhanced trading experience" by providing an "optimally configured desktop screen" that gives access to "unique market views, facilitating a route to the liquidity for market participants to view all their venues aggregated into one stack." (Compl. ¶20). The platform is also alleged to integrate with third-party services for functions like chart analytics. (Compl. ¶21).

No probative visual evidence provided in complaint.

IV. Analysis of Infringement Allegations

The complaint references but does not include Appendices A, B, and C, which purportedly contain detailed claim charts. The infringement analysis is therefore based on the narrative allegations in the body of the complaint.

'886 Patent Infringement Allegations Summary

The complaint alleges that the Accused Instrumentalities directly infringe at least claim 1 of the ’886 Patent. (Compl. ¶¶25-27). The infringement theory posits that Defendant's platform subscribes to continuous input data streams (e.g., real-time market data) in various publisher-specific formats, executes query operations on that data directly and continuously without intermediate storage, and publishes the results to subscribers (e.g., users of the Joule platform) in a different, subscriber-specific format. (Compl. ¶26). The complaint does not, however, provide specific factual support for how the accused platform performs these steps, deferring the details to the unattached Appendix A. (Compl. ¶27).

'927 Patent Infringement Allegations Summary

The complaint alleges that the Accused Instrumentalities directly infringe at least claim 19 of the ’927 Patent. (Compl. ¶¶39-41). The narrative theory is that the platform implements the claimed method by defining "publication schemas" to normalize data from heterogeneous sources (e.g., different energy markets and venues), generating non-permanent data instances, and then using an "integration schema" to combine the data, resolve conflicts, and form a unified "target" view for the user. (Compl. ¶40). This allegedly corresponds to the Joule platform's function of aggregating market data into a single, unified view. (Compl. ¶20). Specifics of this theory are deferred to the unattached Appendix B. (Compl. ¶41).

Identified Points of Contention

• Scope Questions: A primary question for the ’886 Patent is whether the accused platform's processing meets the negative limitation "without storing the continuous input data streams in any data structure." The use of modern in-memory buffering or caching techniques by the accused system could create a factual dispute over whether such transient storage constitutes "storing" in a "data structure" as prohibited by the claim.
• Technical Questions: For the ’927 Patent, a key point of contention may be whether the accused platform's aggregation of data sources involves the formal, structured steps of creating distinct "publication schemas" and an "integration schema" that actively "analyze[s] conflicts," as required by claim 19. The infringement case may depend on whether Defendant’s data aggregation is merely a functional equivalent or if it maps directly to the specific architecture claimed in the patent.

V. Key Claim Terms for Construction

Term from '886 Patent, Claim 1: "without storing the continuous input data streams in any data structure"

• Context and Importance: This negative limitation is central to the infringement analysis for the ’886 patent. If the accused system is found to use any form of intermediate storage that falls within the scope of this term prior to query execution, there would be no literal infringement. Practitioners may focus on this term because it distinguishes the invention from conventional database-centric systems and sets a high technical bar.
• Intrinsic Evidence for a Broader Interpretation: A party arguing for a broader scope of infringement (i.e., a narrower reading of the limitation) may point to the specification's focus on contrasting the invention with traditional databases that store data "on disk." (’886 Patent, col. 2:10-13). This could support an argument that "storing" should be limited to persistent or semi-persistent storage, and not encompass the transient, in-memory buffering inherent in most high-performance network applications.
• Intrinsic Evidence for a Narrower Interpretation: The claim language itself is absolute ("any data structure"). A party arguing for a narrower scope of infringement may contend that any organized representation of data in memory, such as a queue or buffer, qualifies as a "data structure." The patent does not appear to contain an explicit definition or disclaimer that would exclude such common, transient structures from the scope of this limitation.

Term from '927 Patent, Claim 19: "integration schema"

• Context and Importance: The existence and function of an "integration schema" is a core element of the asserted method claim in the ’927 patent. The infringement question will turn on whether the accused system's method for combining data from different sources can be properly characterized as establishing and using such a schema.
• Intrinsic Evidence for a Broader Interpretation: Claim 19 defines the term functionally as "defining a set of integration rules for combining data" and "being used to analyze conflicts." (’927 Patent, Claim 19). A plaintiff may argue this language does not require a specific implementation, and that any software logic that combines data and handles inconsistencies meets the definition.
• Intrinsic Evidence for a Narrower Interpretation: The patent's detailed description and figures suggest a formal, multi-step process for creating the schema, which includes distinct "detecting conflicts" and "processing conflicts" stages. (e.g., ’927 Patent, Fig. 3, steps 35-37). A defendant may argue that the term requires a discrete, definable object with explicit rules for conflict resolution, not merely an algorithm that implicitly merges data streams.

VI. Other Allegations

Indirect Infringement

The complaint alleges both induced infringement (§ 271(b)) and contributory infringement (§ 271(c)) for all three asserted patents. (Compl. ¶¶31-37, 45-51, 59-65). The inducement allegations are based on Defendant’s marketing materials, user guides (specifically the "Joule Direct User Guide"), and training services, which allegedly instruct customers on how to use the accused products in an infringing manner. (Compl. ¶¶34, 48, 62). The contributory infringement allegations assert that the accused functionality is especially made for infringement and is not a staple article of commerce suitable for substantial noninfringing use. (Compl. ¶¶36, 50, 64).

Willful Infringement

Willfulness is alleged for all three patents based on knowledge obtained "at least as early as the filing of this Complaint." (Compl. ¶¶29, 43, 57). This establishes a basis for seeking enhanced damages for any infringement occurring after Defendant received notice of the lawsuit.

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

• A core issue will be one of operational fidelity: Does the accused Trayport platform, a commercial trading system, actually perform the highly specific, and in some cases negative, technical steps recited in the patents? For the '886 patent, this centers on whether its real-time processing occurs "without storing" data in "any data structure," while for the '927 patent, it rests on whether its data aggregation methods can be characterized as the formal creation of "publication" and "integration" schemas.
• The outcome of the case may also depend on definitional scope: How the court construes foundational claim terms will be critical. A narrow construction of terms like "without storing" ('886 patent) and "integration schema" ('927 patent), limiting them strictly to the patent's specific embodiments, could pose a significant challenge to the infringement allegations. Conversely, a broader, more functional interpretation could bring a wider range of modern data-handling architectures within the patents' reach.