DCT
2:24-cv-03951
Sandpiper CDN LLC v. Google LLC
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Sandpiper CDN, LLC (Delaware)
- Defendant: Google LLC (Delaware)
- Plaintiff’s Counsel: Shook, Hardy & Bacon LLP.
- Case Identification: [Sandpiper CDN, LLC](https://ai-lab.exparte.com/party/sandpiper-cdn-llc) v. [Google LLC](https://ai-lab.exparte.com/party/google-llc), 2:24-cv-03951, C.D. Cal., 05/10/2024
- Venue Allegations: Plaintiff alleges venue is proper in the Central District of California because Google maintains regular and established places of business in the district and has committed, and continues to commit, acts of patent infringement within the district.
- Core Dispute: Plaintiff alleges that Defendant’s content delivery network services, including Google Cloud CDN and YouTube, infringe six patents related to foundational content delivery network (CDN) technology.
- Technical Context: Content Delivery Networks are a critical component of modern internet infrastructure, responsible for efficiently distributing high-bandwidth content such as streaming video and complex webpages to a global user base by caching content at geographically distributed servers.
- Key Procedural History: The complaint asserts that the patented technology was originally developed by Sandpiper Networks, Inc. in the mid-to-late 1990s. The patents and associated CDN assets were subsequently acquired by Digital Island, Inc. and later by Level 3 Communications in 2007. Level 3 sold the asserted patents to the current plaintiff, Sandpiper CDN, LLC, on March 29, 2024, shortly before this lawsuit was filed.
Case Timeline
| Date | Event |
|---|---|
| 1996-05-24 | Sandpiper team conceives techniques for delivering streaming resources. |
| 1998-02-10 | Earliest priority date for U.S. Patent No. 8,478,903. |
| 2001-10-18 | Earliest priority date for U.S. Patent No. 9,021,112. |
| 2002-09-20 | Earliest priority date for U.S. Patent No. 8,645,517. |
| 2007-01-01 | Level 3 acquires assets of Sandpiper Networks and Digital Island. |
| 2008-04-04 | Earliest priority date for U.S. Patent No. 10,924,573. |
| 2008-11-12 | Earliest priority date for U.S. Patent Nos. 8,595,778 and 8,719,886. |
| 2013-07-02 | U.S. Patent No. 8,478,903 issues. |
| 2013-11-26 | U.S. Patent No. 8,595,778 issues. |
| 2014-02-04 | U.S. Patent No. 8,645,517 issues. |
| 2014-05-06 | U.S. Patent No. 8,719,886 issues. |
| 2015-04-28 | U.S. Patent No. 9,021,112 issues. |
| 2021-02-16 | U.S. Patent No. 10,924,573 issues. |
| 2023-01-01 | Level 3 decides to exit the CDN market. |
| 2024-03-29 | Level 3 sells the Asserted Patents to Plaintiff Sandpiper CDN. |
| 2024-05-10 | Complaint Filing Date. |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 8,478,903 - "Shared Content Delivery Infrastructure"
The Invention Explained
- Problem Addressed: The patent addresses the challenge of a content provider using a third-party CDN to deliver its web content without losing its brand identity. Specifically, it seeks to solve the problem of how a CDN can serve content for multiple providers while still allowing each provider's unique internet domain to be displayed in the end user's browser (Compl. ¶33; ’903 Patent, col. 1:15-22). The patent also addresses reducing server load and traffic to the providers' origin servers (Compl. ¶34).
- The Patented Solution: The invention proposes a content delivery system with a shared "repeater server" (an early term for a CDN edge server) that serves resources for multiple, distinct content providers. The system associates different "alias names" with resources from different providers. When a user requests a resource, the repeater server uses a "table listing origin servers" to determine which provider's origin server is associated with the requested resource and its alias, enabling the CDN to retrieve and serve the correct content (Compl. ¶35; ’903 Patent, Abstract).
- Technical Importance: This technology allowed CDN customers to leverage the performance and scalability of a shared CDN infrastructure while maintaining their own domain names, a key aspect of online branding (Compl. ¶35).
Key Claims at a Glance
- The complaint asserts independent claim 28 (Compl. ¶57).
- Essential elements of claim 28 include:
- A method in a content delivery system for delivering content for a plurality of origin servers.
- The content delivery system comprising at least one shared repeater server operable to replicate resources.
- Associating the repeater server with a first alias name for a first resource on a first origin server.
- Associating the repeater server with a second alias name for a second resource on a second origin server.
- Providing a table listing origin servers having content authorized for delivery.
- The repeater server analyzing, using the table and an alias name from a client request, to determine the origin server for the requested resource.
- The complaint does not explicitly reserve the right to assert dependent claims for this patent.
U.S. Patent No. 8,595,778 - "User Authentication in a Content Delivery Network"
The Invention Explained
- Problem Addressed: The patent addresses challenges in managing digital rights for multimedia content delivered via a CDN. Specifically, it confronts the issue of how to authorize or deny user requests for content, such as video streams, based on factors like the user's geographic location or account status (Compl. ¶37; ’778 Patent, col. 1:15-26).
- The Patented Solution: The invention describes a method where a request for a video stream is received across a network. This triggers a query to a "subscription database." Based on the reply from that database, the system either initiates delivery of the video stream to the authorized user or transmits a notification to the user that they are not authorized (Compl. ¶37; ’778 Patent, Abstract). The specification describes using a "subscriber verification table" to store information about authorized subscribers and their sessions (’778 Patent, col. 2:1-10).
- Technical Importance: This method provided content providers with a technical solution to protect their media assets and enforce licensing agreements, such as geo-fencing content to specific regions, over a distributed network (Compl. ¶38).
Key Claims at a Glance
- The complaint asserts independent claim 1 (Compl. ¶60).
- Essential elements of claim 1 include:
- A method for authorizing delivery of a video stream to an end user.
- Receiving a request across a network from the end user for delivery of the video stream.
- Querying a subscription database associated with a content publisher.
- Processing a reply from the subscription database to determine if the end user is authorized.
- Performing at least one of: transmitting a notification that the user is not authorized, or initiating delivery of the video stream.
- The complaint does not explicitly reserve the right to assert dependent claims for this patent.
U.S. Patent No. 8,645,517 - "Policy-Based Content Delivery Network Selection"
- Technology Synopsis: The patent addresses the problem of directing network traffic for content requests across multiple CDNs based on various policies (Compl. ¶¶40-41). The solution involves using Domain Name Server (DNS) technology to resolve requests based on configurable network traffic rules that can account for factors like server availability, geolocation, load, and latency, and can be set using a graphical user interface with a decision tree structure (Compl. ¶¶41-42).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶62).
- Accused Features: Google's CDN, including its use of Google Cloud DNS to implement policy-based traffic direction via a graphical user interface (Compl. ¶¶52, 62).
U.S. Patent No. 8,719,886 - "Dynamic Processing of Streamed Content"
- Technology Synopsis: The patent addresses the problem of modifying streamed content, such as video, while it is being delivered over a CDN (Compl. ¶43). The solution involves detecting a "trigger signal" within the video stream, processing that signal to determine if modification is needed (e.g., by querying a data repository related to a programming schedule), and then modifying the stream, such as by injecting selected content (Compl. ¶¶44, 64).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶64).
- Accused Features: Google's YouTube and YouTube TV services, which allegedly deliver video streams containing trigger signals used to modify the stream, for example by inserting advertisements based on a programming schedule (Compl. ¶¶53, 64).
U.S. Patent No. 9,021,112 - "Content Request Routing and Load Balancing For Content Distribution Networks"
- Technology Synopsis: The patent addresses the need to respond to content requests quickly and efficiently from multiple servers in a CDN, improving upon anycast DNS to reduce bandwidth and latency issues (Compl. ¶¶46-47). The solution uses DNS technology and a common anycast address shared by a plurality of DNS servers, which routes a content request to the server closest to the user in terms of network distance (Compl. ¶¶47, 66).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶66).
- Accused Features: Google's CDN, which allegedly uses a plurality of DNS servers (like Google Cloud DNS) sharing a common anycast address to resolve hostnames and direct client requests to an appropriate content server (Compl. ¶¶52, 66).
U.S. Patent No. 10,924,573 - "Handling Long-Tail Content in a Content Delivery Network (CDN)"
- Technology Synopsis: The patent addresses problems with handling large libraries of content where only a small fraction is popular ("short head") and the rest is accessed infrequently ("long tail") (Compl. ¶¶49-50). The solution is a tiered server system where a first-tier server handles a request; if it lacks the content, it determines the content's popularity. If popular, it retrieves and caches the content; if not, it redirects the request to a higher-tier server, thus conserving cache space on edge servers (Compl. ¶50).
- Asserted Claims: Independent claim 1 is asserted (Compl. ¶68).
- Accused Features: Google's CDN, which allegedly uses a tiered server system that accesses a "popularity service" to determine a popularity designation for a resource and processes redirects based on that designation (Compl. ¶68).
III. The Accused Instrumentality
Product Identification
- The accused instrumentalities are Google's content delivery network services, collectively referred to as "Google CDN" (Compl. ¶27). This includes specific services such as Google Cloud CDN, Google Media CDN, Google Cloud DNS, YouTube, and YouTube TV (Compl. ¶28).
Functionality and Market Context
- The complaint alleges that Google's CDN is an in-house system built by Google that incorporates the patented technologies (Compl. ¶27). The accused functionality includes providing cached content from edge servers to reduce latency, as highlighted in a screenshot describing "Cache hits and cache misses" (Compl. p. 15; Compl. ¶51). It also allegedly uses a global network of "anycast name servers" via its Cloud DNS service to provide low-latency DNS resolution (Compl. p. 17; Compl. ¶52). The complaint further alleges that the YouTube and YouTube TV platforms use the Google CDN to deliver video content, including streaming with dynamically inserted ads and providing access to content based on user subscriptions (Compl. ¶¶53-54). The complaint positions Google CDN as a direct competitor to the CDN services offered by the Plaintiff's predecessor, Level 3 (Compl. ¶29).
IV. Analysis of Infringement Allegations
’903 Patent Infringement Allegations
| Claim Element (from Independent Claim 28) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| a method in a content delivery system operative in a computer network for delivering content to client machines, the computer network comprising a plurality of origin servers, each of said origin servers having resources associated therewith, and the content delivery system comprising at least one shared repeater server operable to replicate resources associated with the plurality of origin servers | Google provides a CDN for delivering content from multiple origin servers to client machines, using shared servers to replicate resources. | ¶58 | col. 1:15-22 |
| associating the at least one repeater server with a first alias name, wherein requests for a first resource, located on a first origin server, are directed based at least in part on the first alias name | Google's CDN associates its repeater servers with a first alias name to direct requests for a first resource from a first origin server. | ¶58 | col. 8:1-5 |
| associating the at least one repeater server with a second alias name, wherein requests for a second resource located on a second origin server are directed to the at least one repeater server for delivery of the second resource | Google's CDN associates the same repeater servers with a second alias name to direct requests for a second resource from a second origin server. | ¶58 | col. 8:6-10 |
| providing a table listing origin servers having content located thereon, wherein said content is authorized for delivery to client machines via the at least one shared repeater server wherein the origin servers comprise a first and second origin server | Google provides a table that lists the origin servers and their content, which is used to authorize delivery via the shared repeater servers. | ¶58 | col. 2:22-26 |
| the at least one repeater server...constructed and adapted to analyze, using the table and the alias name received with a client request to determine the origin server associated with the requested resource | Google's repeater server is constructed to use the table and the received alias name to determine the correct origin server for a requested resource. | ¶58 | col. 2:27-31 |
’778 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| A method for authorizing delivery of a video stream to an end user | Google provides a method for authorizing delivery of video streams to users of services like YouTube and YouTube TV. | ¶60 | col. 1:15-18 |
| receiving a request across a network from the end user for delivery of the video stream | YouTube TV receives a request from an end user for delivery of a video stream, as shown in a screenshot of the YouTube TV interface allowing users to select an event to stream. (Compl. p. 19; Compl. ¶55). | ¶60 | col. 16:6-8 |
| querying a subscription database associated with a content publisher | Google's system queries a database of subscribers to determine authorization for content, such as for YouTube TV, which requires users to subscribe to access certain video streams. (Compl. ¶54). | ¶60 | col. 1:49-51 |
| processing a reply from the subscription database to determine whether the end user has authorization to receive delivery of the video stream | Google's system processes the information returned from its user/subscription database to determine if access should be granted. | ¶60 | col. 1:52-55 |
| performing at least one of: (i) transmitting a notification to the end user that the end user is not authorized...and (ii) initiating delivery of the video stream to the end user... | Based on the authorization check, Google's system either initiates delivery of the stream to the user or denies access. | ¶60 | col. 1:55-61 |
Identified Points of Contention
- Scope Questions: A central question for the ’903 patent, which claims priority to 1998, will be whether its term "table listing origin servers" can be construed to cover the complex, likely distributed, and dynamic routing and mapping systems used in Google's modern cloud-based CDN architecture. For the ’778 patent, a question is whether the term "subscription database" is limited to paid subscription services or if it can read on any user account database that governs content access rights, including for ad-supported or geographically restricted content.
- Technical Questions: For the ’903 patent, the complaint provides a conclusory allegation of a "table." A key factual question will be what specific component or data structure within Google's CDN architecture performs the function of this claimed "table." For the ’778 patent, a question will be what evidence demonstrates that Google's systems perform the specific sequence of "querying," "processing a reply," and then "initiating delivery or transmitting a notification" as distinct, ordered steps required by the claim.
V. Key Claim Terms for Construction
The Term: "table listing origin servers" (from ’903 Patent, claim 28)
- Context and Importance: This term is critical because infringement hinges on whether Google's modern, distributed CDN architecture contains a structure that meets this limitation. Practitioners may focus on this term because the complaint alleges its existence (Compl. ¶58) but does not identify a specific technical implementation, suggesting a potential dispute over whether Google's dynamic routing maps or other data structures constitute a "table" as understood in the patent.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification describes the system in functional terms, stating a need to "determine a different server (a 'repeater') to process those requests" and that "The selection of the repeater can be made dynamically" (’903 Patent, col. 2:61-65). This functional language may support an argument that any mechanism that dynamically maps requests to origin servers, regardless of its specific data structure, falls within the scope.
- Evidence for a Narrower Interpretation: The claim language itself uses the specific word "table." The abstract refers to a "table... that can be consulted to determine a content provider associated with a particular resource" (’903 Patent, Abstract). This may support an interpretation that requires a more literal, discrete data structure, akin to a lookup table, rather than a distributed or algorithmic routing logic system.
The Term: "subscription database" (from ’778 Patent, claim 1)
- Context and Importance: The definition of this term will determine the breadth of accused services. A narrow definition limited to paid subscriptions might restrict the claim's applicability, while a broader definition could encompass a wider range of Google's services that control access based on user account status, geography, or other criteria.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The patent's background describes the need to manage digital rights generally, including for users who are not authorized due to "location or account status" (’778 Patent, col. 1:21-23), not just payment status. This could support a reading where any database that stores user-specific authorization rules (geographic, age, etc.) is a "subscription database."
- Evidence for a Narrower Interpretation: The claim uses the specific term "subscription." The patent repeatedly discusses a "subscriber verification table" with "entries regarding session information" for "authorized subscribers" (’778 Patent, col. 2:1-10). This language, combined with the common understanding of "subscription," could support an argument that the term is limited to databases managing formal, opt-in subscriber relationships, such as for a paid service like YouTube TV.
VI. Other Allegations
Willful Infringement
- The complaint does not use the word "willful," but it requests enhanced damages under 35 U.S.C. § 284 and attorneys' fees for an "exceptional case" under § 285 (Compl., Prayer for Relief ¶¶ B, C, E). The factual basis alleged is that Google built its CDN and began competing with Plaintiff's predecessors after the "foundational CDN technology" was pioneered and patented, thereby "misappropriating years of research and investments" without a license (Compl. ¶¶26-29).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of temporal scope and construction: Can claim terms rooted in the context of late-1990s and early-2000s internet architecture (e.g., the '903 patent's "table listing origin servers") be construed to read on the highly distributed, virtualized, and algorithmically complex infrastructure of Google's modern cloud CDN?
- A key evidentiary question will be one of functional mapping: How will the Plaintiff demonstrate that the multifaceted and integrated processes of Google's services like YouTube and YouTube TV perform the specific, discrete steps recited in the asserted method claims (e.g., the '778 patent's sequence of "querying a subscription database," "processing a reply," and "initiating delivery")? The case may turn on whether these claimed steps can be clearly mapped to distinct operations within the accused systems or are instead emergent functions of a more holistic process.