DCT

3:24-cv-01795

Valtrus Innovations Ltd v. Google LLC

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 3:24-cv-01795, N.D. Tex., 12/18/2024
  • Venue Allegations: Venue is alleged to be proper in the Northern District of Texas because Google maintains a regular and established place of business in the district, including a 260,000 square foot data center in Midlothian, and has committed acts of infringement there.
  • Core Dispute: Plaintiffs allege that Defendant’s Google Search and Google Cloud services infringe three patents related to merging search engine results, hardware-based CPU utilization metering, and ranking hypertext search results.
  • Technical Context: The technologies at issue relate to foundational aspects of internet search result aggregation and the underlying hardware management for large-scale cloud computing services.
  • Key Procedural History: The complaint alleges that the '704 Patent overcame prior art rejections during prosecution by adding limitations regarding the dynamic nature of result rankings. The '809 Patent prosecution allegedly emphasized its hardware-based approach as novel over prior software methods. The '604 Patent issued after a favorable decision on appeal at the Patent Trial and Appeal Board (PTAB), which the complaint asserts found the invention novel and non-obvious.

Case Timeline

Date Event
1999-10-15 ’604 Patent Priority Date
2001-08-27 ’704 Patent Priority Date
2002-07-23 ’809 Patent Priority Date
2004-04-27 ’704 Patent Issue Date
2004-11-09 ’809 Patent Issue Date
2007-05-01 Alleged launch of Google’s “Universal Search” feature (representative date)
2008-03-18 ’604 Patent Issue Date
2012-06-26 Alleged date of Google’s knowledge of ’604 Patent
2013-01-01 Alleged date of Google’s knowledge of ’704 Patent
2021-04-14 Alleged date of Google’s knowledge of ’809 Patent
2024-12-18 Complaint Filing Date

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

U.S. Patent No. 6,728,704 - “Method and apparatus for merging result lists from multiple search engines,” issued April 27, 2004

The Invention Explained

  • Problem Addressed: At the time of the invention, merging result lists from multiple search engines was computationally intensive because it required examining and ranking every entry from every list, which could "nullify any advantage gained by operating multiple search engines at the same time" (’704 Patent, col. 2:53-56; Compl. ¶16).
  • The Patented Solution: The invention proposes a more efficient method where only a subset of entries from each result list is evaluated. A "representative value" (e.g., an average score) is assigned to each list based on its subset. The complete lists are then merged into a single ranked list based on these representative values, which can be dynamically adjusted during the merging process, thereby reducing the total number of calculations required (’704 Patent, Abstract; col. 3:20-24; Compl. ¶¶17, 20).
  • Technical Importance: This approach aimed to improve the speed and efficiency of meta-search engines by reducing computational overhead, allowing for faster aggregation of results from numerous sources without sacrificing accuracy (Compl. ¶18).

Key Claims at a Glance

  • Independent Claim 1 (method):
    • transmitting a query to a set of search engines;
    • receiving a result list from each search engine;
    • selecting a subset of entries from each result list;
    • assigning a scoring value to each entry in the subset;
    • assigning to each subset a representative value according to the scoring values of its entries;
    • producing a merged list of entries in a predetermined manner based on the representative value, wherein the representative value varies in accordance with the predetermined manner.
  • Independent Claim 12 (method): A similar method claim that recites "characterizing said subset in accordance with a representative value" and merging based on that value.
  • The complaint also asserts dependent claims 2-6 and 13, and reserves the right to assert others (Compl. ¶¶25, 46, 49, 66).

U.S. Patent No. 6,816,809 - “Hardware based utilization metering,” issued November 9, 2004

The Invention Explained

  • Problem Addressed: Metering processor utilization for billing purposes was typically done via software running within an operating system. This posed "significant challenges" in partitioned hardware environments where multiple, isolated operating systems might be running, as the metering software would lack visibility into other systems (’809 Patent, col. 1:16-27; Compl. ¶79).
  • The Patented Solution: The patent describes a hardware-based device for metering utilization. The device includes an "idle indicator" coupled to the processor, a counter that receives inputs from the idle indicator and a system clock, and a "data usage provider" capable of providing the counter's value. This hardware-level approach is designed to function independently of any specific operating system running on the processor (’809 Patent, Abstract; col. 4:1-23; Compl. ¶¶80, 89-93).
  • Technical Importance: A hardware-based solution sought to provide a more reliable and universal method for usage metering in complex, multi-tenant computing environments, avoiding the pitfalls of software-based agents that were OS-dependent and could fail (’809 Patent, col. 3:5-18; Compl. ¶84).

Key Claims at a Glance

  • Independent Claim 1 (apparatus):
    • an idle indicator coupled to a processor to receive an indication when the processor is in a first state;
    • a counter coupled to the idle indicator and a system clock to receive time and state data and generate a counter value;
    • a data usage provider coupled to the counter, capable of providing the counter value.
  • Independent Claim 13 (method):
    • determining when any of a plurality of processors is busy;
    • providing a busy indication to a counter;
    • receiving at the counter a measure of computer system time;
    • incrementing a counter value based on the busy indication and time;
    • maintaining the counter value.
  • The complaint also asserts dependent claims 2, 3, 5, 6, 9, and 14, among others (Compl. ¶¶83, 95, 98, 101, 104, 107, 124).

Multi-Patent Capsule: U.S. Patent No. 7,346,604 - “Method for ranking hypertext search results by analysis of hyperlinks from expert documents and keyword scope,” issued March 18, 2008

  • Technology Synopsis: The patent addresses the challenge of identifying authoritative search results and filtering out "spam pages" created to mislead search engines (Compl. ¶139). The solution involves a two-phase process: first, identifying a set of "expert documents" in a pre-processing step, defined as pages that link to many non-affiliated pages on a topic; and second, when a query is received, ranking target documents based on the endorsements (i.e., hyperlinks) from these pre-identified and ranked experts (’604 Patent, Abstract; Compl. ¶142).
  • Asserted Claims: Independent claim 1 and dependent claims 2, 3, and 20 (Compl. ¶¶148, 164, 167, 170).
  • Accused Features: Google Search is accused of using this method, particularly through algorithms like "Hilltop," which allegedly build a "special index of expert documents" to determine authority and rank search results (Compl. ¶¶152-153). The complaint references a Google article on ranking that describes using links from "other prominent websites" as a signal of trust and authority (Compl. ¶155).

III. The Accused Instrumentality

Product Identification

The complaint names two primary accused instrumentalities: Google Search (accused of infringing the ’704 and ’604 patents) and Google Cloud (accused of infringing the ’809 patent) (Compl. ¶¶29, 86, 148).

Functionality and Market Context

  • Google Search: The complaint alleges that Google Search, upon receiving a query, returns a single merged list of results from its multiple "vertical search services" such as Web, Maps, News, Images, etc. (Compl. ¶¶27, 31). This "Universal Search" feature is alleged to blend results to provide the most relevant information (Compl. ¶27). The complaint further alleges that Google Search uses algorithms to pre-identify authoritative or "expert" documents and then uses links from those documents to rank final search results (Compl. ¶¶152-153, 155).
  • Google Cloud: The complaint alleges that Google Cloud services are powered by CPUs that include hardware performance counters, such as MPERF and APERF, which increment only when a processor core is in a busy state (e.g., C0 state) (Compl. ¶¶88, 90). The data from these counters is allegedly made available through Google Cloud's monitoring services to provide metrics on CPU utilization (Compl. ¶94).

IV. Analysis of Infringement Allegations

’704 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
transmitting a query to a set of search engines A user query is transmitted to Google Search, which in turn transmits the query to its various search engines (e.g., Web, Maps, News) (Compl. ¶33). ¶32-33 col. 5:1-5
receiving in response to said query a result list from each search engine Google Search receives result lists from each of its search services, such as a list of locations from Maps and a list of articles from News (Compl. ¶35). The complaint provides a screenshot showing a single merged results page for the query "in-n-out burger" that combines web, map, and news results (Compl. ¶31). ¶34-35 col. 5:31-33
selecting a subset of entries from each result list to form a set of selected entries For a given query, only a small subset of the millions or billions of possible results is selected for display to the user (Compl. ¶37). The complaint includes a screenshot indicating "About 675,000,000 results" were found, from which a small number are displayed (Compl. ¶37). ¶36-37 col. 5:44-48
assigning to each selected entry of said set of selected entries a scoring value according to a scoring function Entries are displayed in order of relevance, which is based on a scoring value assigned according to metrics like user distance for Maps results (Compl. ¶39). ¶38-39 col. 5:57-59
assigning to each subset a representative value according to the scoring values assigned to its entries Google Search allegedly orders the subsets themselves (e.g., Maps results vs. News results) based on a representative value that reflects the subset's relevance to the query (Compl. ¶41). ¶40-41 col. 5:61-64
producing a merged list of entries in a predetermined manner based on the representative value assigned to each result list, wherein the representative value varies in accordance with predetermined manner Google Search produces a single merged list where result groups appear in an order determined by their representative value, which varies by query (e.g., Maps results are prioritized for location-based queries) (Compl. ¶¶43, 45). The complaint includes a screenshot showing how different groups of results (e.g., groups of 3 results for "hospital" vs. "target") are displayed based on the user's query (Compl. ¶45, p. 18). ¶42-45 col. 6:5-28; col. 8:17-18
  • Identified Points of Contention:
    • Scope Questions: A central question may be whether Google's internal vertical search properties (e.g., Maps, News, Images) constitute "multiple search engines" as contemplated by the patent, which was filed in an era where meta-search often involved querying distinct, unaffiliated, third-party engines. The defense may argue these are merely different features of a single, integrated search engine.
    • Technical Questions: The complaint alleges that Google assigns a "representative value" to each subset of results (e.g., the Maps block). A factual question will be whether Google's internal ranking and presentation logic for its blended results operates in a way that maps onto the specific method of calculating and dynamically decrementing a "representative value" as described in the patent's embodiments (e.g., ’704 Patent, col. 7:5-20).

’809 Patent Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
an idle indicator coupled to a processor, wherein the idle indicator receives an indication when the processor is in a first state The processors in Google Cloud allegedly use a hardware idle indicator that starts and stops hardware performance counters (MPERF/APERF) based on whether the processor core is in a first state, such as a C0 (busy) state (Compl. ¶90). ¶89-90 col. 4:36-46
a counter coupled to the idle indicator and coupled to a system clock... and generates a counter value indicative of time the processor is in the first state Hardware performance counters (e.g., MPERF/APERF) in Google Cloud processors allegedly increment only when the core is in a busy state, generating a value indicative of the time spent in that state (Compl. ¶92). ¶91-92 col. 4:1-7
a data usage provider coupled to the counter, wherein the data usage provider is capable of providing the counter value Google Cloud provides CPU utilization metrics to customers, which allegedly constitutes a "data usage provider" making the counter value available (Compl. ¶94). The complaint includes a screenshot of a "Google Cloud metrics" table for "CPU utilization" (Compl. ¶94). ¶93-94 col. 4:8-14
  • Identified Points of Contention:
    • Scope Questions: Practitioners will focus on whether the allegedly infringing system components meet the definitions of the claimed hardware elements. For example, does a processor's internal state (e.g., C0) and the associated logic that gates performance counters constitute a distinct "idle indicator" as claimed, or is it an inseparable part of the CPU core itself? Similarly, does Google's software-based monitoring infrastructure that exposes counter data qualify as a hardware "data usage provider" as described in the patent?
    • Technical Questions: A key technical question will be the degree to which the accused system is truly "hardware based." The patent was prosecuted to overcome prior art that was "software running within the computer system's operating system" (Compl. ¶78). The infringement analysis may turn on whether Google's use of readable hardware performance counters that are accessed and reported by software is functionally equivalent to the hardware-centric architecture described in the patent, or if it is closer to the software-based methods the patent sought to improve upon.

V. Key Claim Terms for Construction

For the ’704 Patent

  • The Term: "multiple search engines"
  • Context and Importance: The construction of this term is critical. If construed narrowly to mean distinct and separate corporate entities providing search services (as was common for meta-search engines in 2001), Google may argue its integrated vertical search features do not meet this limitation. If construed broadly to mean functionally distinct search systems, regardless of ownership, Plaintiffs' case may be supported. Practitioners may focus on this term because the factual basis of the infringement allegation rests on equating Google's internal services (Maps, News, etc.) with the "multiple search engines" of the claim.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent specification describes federated, peer-to-peer, and meta-search engines, but does not explicitly limit the term "search engines" to third-party entities. It speaks of engines that "search a group of networked computers" (’704 Patent, col. 1:41-43), which could arguably include internal, networked systems.
    • Evidence for a Narrower Interpretation: The background section contrasts different types of engines, including "web meta-search engines" that "farm a WWW search out to other public web search engines" (’704 Patent, col. 2:29-33). This context may suggest an understanding of separate, public-facing entities.

For the ’809 Patent

  • The Term: "idle indicator"
  • Context and Importance: This term's definition is central to determining if the accused system is truly "hardware based" as claimed. The complaint alleges that hardware performance counters are "stopped and started by a hardware idle indicator" (Compl. ¶90). The dispute will likely focus on whether a processor's internal power state logic (which controls the counters) is the claimed "idle indicator" or if the patent requires a physically or logically distinct hardware component.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent is not overly specific on the structure of the indicator, stating it "is capable of providing either an 'idle' indication or a 'not-idle'/'busy' indication" (’809 Patent, col. 3:51-53), which could encompass a readable processor state.
    • Evidence for a Narrower Interpretation: The specification discusses embodiments where the indicator is coupled to a specific "pin on the CPU chip" or involves an "externally visible register" set by the OS (’809 Patent, col. 4:39-54). This suggests a component that is distinct from the core processing logic and potentially accessible or observable from outside the CPU's primary execution flow.

VI. Other Allegations

  • Indirect Infringement:
    • '704 Patent: The complaint alleges Google induces infringement by "offering for public use Google Search with the intent to encourage and facilitate infringing uses" (Compl. ¶70).
    • '809 Patent: The complaint alleges inducement by "using, selling, and offering for sale Google Cloud products which rely on processors having the claimed counters" with the intent to have customers use the infringing process (Compl. ¶128).
    • '604 Patent: The complaint alleges inducement by offering Google Search for public use (Compl. ¶174).
  • Willful Infringement:
    • '704 Patent: Willfulness is alleged based on Google's alleged knowledge since at least January 1, 2013, derived from its own patents citing the ’704 Patent as a reference and from hiring two of the patent's co-inventors, one of whom became a Senior Vice President responsible for Google Search (Compl. ¶69).
    • '809 Patent: Willfulness is alleged based on Google's knowledge since at least April 14, 2021, when Plaintiffs allegedly sent Google a notice letter identifying the patent and accused products (Compl. ¶127).
    • '604 Patent: Willfulness is alleged based on knowledge since at least June 26, 2012, from Google's own patents citing the ’604 Patent and from the long-term employment of one of its co-inventors, the "inventor of Google News" (Compl. ¶173).

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

This case presents foundational challenges to the operation of modern internet services, raising several key questions for the court:

  • A primary issue will be one of definitional scope: can the term "multiple search engines," rooted in the 2001 context of distinct web engines, be construed to cover the integrated vertical search features (Maps, News, etc.) of a single, unified service like Google Search? The answer will determine the applicability of the '704 patent to modern search architecture.

  • A second core issue is one of architectural characterization: does Google Cloud's use of software-accessible hardware performance counters (e.g., MPERF/APERF) embody the "hardware based utilization metering device" claimed in the '809 patent, or does it represent a software-centric monitoring system that falls outside the patent's hardware-focused claims?

  • Finally, a key evidentiary question for the '604 patent will be one of algorithmic equivalence: does Google's proprietary method for determining website authority and ranking search results, such as its "Hilltop" algorithm, practice the specific two-phase method of pre-identifying "expert documents" and then performing "target ranking" as claimed in the patent?