DCT
2:23-cv-00024
Winterspring Digital LLC v. Marvell Intl Ltd
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Winterspring Digital LLC (Texas)
- Defendant: Marvell International, Ltd. and Marvell Technology Group Ltd. (Bermuda)
- Plaintiff’s Counsel: Fabricant LLP; Truelove Law Firm, PLLC
- Case Identification: 2:23-cv-00024, E.D. Tex., 01/23/2023
- Venue Allegations: Plaintiff alleges venue is proper because Defendants are not U.S. residents and may therefore be sued in any judicial district.
- Core Dispute: Plaintiff alleges that Defendant’s high-speed networking components, including transceivers and Ethernet adaptors, infringe three patents related to 10-Gigabit Ethernet signal transmission, high-speed frame tagging, and network traffic admission control.
- Technical Context: The technology at issue involves the hardware and methods used to manage and process high-bandwidth data traffic, which is a foundational element of modern enterprise networks, data centers, and cloud infrastructure.
- Key Procedural History: The complaint does not mention any prior litigation, Inter Partes Review (IPR) proceedings, or licensing history related to the patents-in-suit.
A. Case Timeline
| Date | Event |
|---|---|
| 2000-07-28 | ’468 Patent Priority Date |
| 2002-04-08 | ’692 Patent Priority Date |
| 2002-12-20 | ’975 Patent Priority Date |
| 2007-01-16 | ’692 Patent Issue Date |
| 2008-09-02 | ’975 Patent Issue Date |
| 2010-08-10 | ’468 Patent Issue Date |
| 2023-01-23 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
A. U.S. Patent No. 7,164,692 - “Apparatus and Method for Transmitting 10 Gigabit Ethernet LAN Signals Over a Transport System,” issued January 16, 2007
1. The Invention Explained
- Problem Addressed: The patent describes a problem where high-speed Local Area Network (LAN) technologies like Ethernet were limited in the distance they could span. To extend them over long-haul Wide Area Networks (WANs), Ethernet frames had to be encapsulated within complex and costly transport protocols like SONET, adding unnecessary layers and expense (’692 Patent, col. 1:55-2:18, col. 3:1-10). The industry needed a way to transmit native 10-Gigabit Ethernet (10GE) signals over long-haul systems without this inefficient protocol conversion (’692 Patent, col. 4:45-51).
- The Patented Solution: The invention is a transceiver that receives a standard 10GE LAN signal in its native format and transmits it over a long-haul transport system without converting it to a SONET-based standard. The transceiver converts the signal to an internal electrical form, re-clocks and regenerates it to maintain signal integrity, and then re-transmits it onto the transport system, optionally using Forward Error Correction (FEC) to manage errors over long distances (’692 Patent, Abstract; col. 5:18-34). The internal components of the transceiver are illustrated in the patent’s Figure 3 (’692 Patent, Fig. 3).
- Technical Importance: This approach allowed for the extension of native Ethernet networks over metropolitan or long-haul distances, which could simplify network architecture and reduce costs by eliminating the need for separate, expensive WAN-specific routing equipment and complex protocol conversions (’692 Patent, col. 6:9-16).
2. Key Claims at a Glance
- The complaint asserts at least independent claim 10 (Compl. ¶20).
- Essential elements of claim 10, a method claim, include:
- receiving the 10GE LAN client signal transmitted over the transport system;
- converting the 10GE LAN client signal to an intermediate signal;
- recovering clock data from the intermediate signal;
- recovering a data stream from the intermediate signal;
- reconverting the intermediate signal to the 10GE LAN client signal;
- transferring the 10GE LAN client signal to a client system; and
- monitoring the intermediate form with a monitoring device, wherein the monitoring device is a 10GE LAN media access controller.
- The complaint does not explicitly reserve the right to assert dependent claims.
B. U.S. Patent No. 7,420,975 - “Method and Apparatus For High-Speed Frame Tagger,” issued September 2, 2008
1. The Invention Explained
- Problem Addressed: As network speeds increase, network processors become a bottleneck because they are tasked with both processing the data within a packet and parsing the packet to determine its protocol type (e.g., control vs. data). This dual workload prevents them from operating at the full "line speed" of the network connection (’975 Patent, col. 1:30-41).
- The Patented Solution: The patent proposes a dedicated hardware apparatus, or "frame tagger," that offloads the task of protocol identification from the main network processor. This apparatus performs a multi-pass comparison on an incoming packet at line speed, determines its protocol type based on pre-defined values, and applies a "tag." The network processor can then use this simple tag to quickly determine how to handle or route the packet without having to perform the complex analysis itself (’975 Patent, Abstract; col. 2:40-57). The patent’s Figure 4 details the logic blocks for performing this analysis (’975 Patent, Fig. 4).
- Technical Importance: By dedicating specialized hardware to the task of high-speed packet classification, this invention allows network systems to more efficiently process high-volume data streams, enabling network processors to keep pace with increasing line speeds.
2. Key Claims at a Glance
- The complaint asserts at least independent claim 5 (Compl. ¶30).
- Essential elements of claim 5, an apparatus claim, include:
- a network processor interface suitable for coupling to a network processor;
- a central processor interface suitable for coupling to a central processor;
- a protocol determination logic block that compares protocol information in a first pass to produce a first result, and if the first result is positive, compares the protocol information in a second pass to produce a second result; and
- a tag select logic block that applies a tag to the packet, where the tag indicates an "unknown protocol type" if the first result is negative, and directs the packet to either the central processor or network processor interface if the first result is positive.
- The complaint does not explicitly reserve the right to assert dependent claims.
C. U.S. Patent No. 7,774,468 - “Network Traffic Admission Control,” issued August 10, 2010
1. Technology Synopsis
The patent addresses the challenge of managing bandwidth in a Differentiated Services (DiffServ) network, where edge devices typically lack visibility into congestion on the full network path. The invention describes a system where a central "media director" with global network visibility allocates a "real-time bandwidth pool" to an edge device for a specific path. The edge device can then perform local admission control for new sessions against this pre-allocated bandwidth, avoiding the high overhead of querying the entire network for each new connection request (’468 Patent, Abstract; col. 2:28-49).
2. Asserted Claims
The complaint asserts at least independent claim 1 (Compl. ¶40).
3. Accused Features
The complaint alleges that the Marvell FastLinQ 41000 Series Ethernet Adaptor infringes by performing a method of requesting and receiving an allocation of bandwidth as a "real-time bandwidth pool" from a director node and subsequently allocating portions of that pool to establish real-time communication sessions (Compl. ¶¶ 40-41).
III. The Accused Instrumentality
A. Product Identification
The complaint names the Marvell AQS-107 10G transceiver and the Marvell FastLinQ 41000 Series Ethernet Adaptor (Compl. ¶¶ 21, 31, 41).
B. Functionality and Market Context
- The complaint identifies the accused products as components used in network switches, routers, adaptors, controllers, and optical modules (Compl. ¶¶ 16, 19).
- The Marvell AQS-107 10G transceiver is alleged to perform a method of transferring 10GE LAN signals by receiving, converting, re-clocking, and monitoring the signals (Compl. ¶21). A product brief included in the complaint describes the AQS-107 module as delivering up to 10GbE network speeds and using "AQrate technology" to perform physical layer functions (Compl. p. 6).
- The Marvell FastLinQ 41000 Series Ethernet Adaptor is alleged to perform packet tagging and traffic admission control (Compl. ¶¶ 31, 41). A product brief for this series, referenced in the complaint, states that the adaptor supports "virtual LAN (VLAN) tagging" (Compl. p. 9). Another document excerpt describes how a "smart line card" can process a packet, determine its destination, and amend the packet with a "small tag" to direct it through the network fabric (Compl. p. 10). For traffic admission, the complaint points to the adaptor's "switch-independent NIC partitioning (NPAR)" feature, which enables segmentation of a port into multiple partitions with dynamic bandwidth allocation (Compl. p. 13).
IV. Analysis of Infringement Allegations
A. ’692 Patent Infringement Allegations
| Claim Element (from Independent Claim 10) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| receiving the 10GE LAN client signal transmitted over the transport system; | The Marvell AQS-107 10G transceiver receives the 10GE LAN client signal transmitted over the transport system. | ¶21 | col. 15:15-17 |
| converting the 10GE LAN client signal to an intermediate signal; | The accused transceiver converts the 10GE LAN client signal to an intermediate signal. | ¶21 | col. 15:18-19 |
| recovering clock data from the intermediate signal; | The accused transceiver recovers clock data from the intermediate signal. | ¶21 | col. 15:20-21 |
| recovering a data stream from the intermediate signal; | The accused transceiver recovers a data stream from the intermediate signal. | ¶21 | col. 15:22-23 |
| reconverting the intermediate signal to the 10GE LAN client signal; | The accused transceiver reconverts the intermediate signal to the 10GE LAN client signal. | ¶21 | col. 15:24-25 |
| transferring the 10GE LAN client signal to a client system; and | The accused transceiver transfers the 10GE LAN client signal to a client system. | ¶21 | col. 15:26-27 |
| monitoring the intermediate form with a monitoring device, wherein the monitoring device is a 10GE LAN media access controller. | The accused transceiver monitors the intermediate form with a monitoring device, which is a 10GE LAN media access controller. | ¶21 | col. 15:28-30 |
1. Identified Points of Contention
- Technical Question: The complaint alleges that the monitoring device in the accused AQS-107 transceiver is a "10GE LAN media access controller" (MAC), mirroring the claim language (Compl. ¶21). A central question will be what evidence demonstrates that the accused product's monitoring component meets the specific technical definition and function of a 10GE LAN MAC, as described in the patent (e.g., ’692 Patent, col. 10:1-25), rather than a more generic monitoring circuit.
B. ’975 Patent Infringement Allegations
| Claim Element (from Independent Claim 5) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| a network processor interface suitable for coupling to a network processor; | The Marvell FastLinQ 41000 Series Ethernet Adaptor includes a network processor interface. | ¶31 | col. 9:51-52 |
| a central processor interface suitable for coupling to a central processor; | The accused adaptor includes a central processor interface. | ¶31 | col. 9:53-54 |
| a protocol determination logic block to determine a protocol type of data in a packet, wherein the protocol determination logic compares the protocol information in a first pass to predetermined values to procedure a first result and, if the first result is positive, compares the protocol information in a second pass to predetermined values to produce a second result, the first and second results forming a set of results; and | The accused adaptor includes a protocol determination logic block that performs a conditional two-pass comparison of protocol information to produce a set of results. | ¶31 | col. 9:55-63 |
| a tag select logic block to apply a tag to the packet indicating that the packet has an unknown protocol type if the first result is negative and if the first result is positive the packet should be sent to either the central processor interface or the network processor interface based on the set of results. | The accused adaptor includes a tag select logic block that applies a tag based on the results of the two-pass comparison to direct the packet. | ¶31 | col. 10:2-11 |
1. Identified Points of Contention
- Technical Question: The complaint alleges, upon "information and belief," that the accused adaptor performs the specific conditional two-pass comparison required by the claim (Compl. ¶31). The infringement analysis may focus on what evidence supports this particular logical operation, as the marketing materials cited in the complaint describe "VLAN tagging" but do not explicitly detail a two-pass comparison method (Compl. p. 9).
- Scope Questions: A question for the court may be whether functionality like "VLAN tagging" falls within the scope of the claimed "protocol determination logic block." The analysis will likely explore whether the accused product's operation matches the claim's requirement for a hierarchical comparison where the result of the first pass dictates whether a second pass occurs.
V. Key Claim Terms for Construction
A. ’692 Patent, Claim 10
- The Term: "monitoring the intermediate form with a monitoring device, wherein the monitoring device is a 10GE LAN media access controller"
- Context and Importance: This limitation defines a specific component responsible for monitoring. The infringement case may depend on whether the component used for monitoring in the accused AQS-107 transceiver can be properly characterized as a "10GE LAN media access controller" (MAC).
- Intrinsic Evidence for a Broader Interpretation: A party could argue that any component performing the functions of a MAC as described in the specification—such as monitoring "packet data, idle, preamble" and performing "cyclic redundancy checks (CRC)" (’692 Patent, col. 10:3-8)—satisfies this limitation, regardless of its formal name.
- Intrinsic Evidence for a Narrower Interpretation: The specification repeatedly and specifically identifies this component as "MAC 312" and describes it as a "standard IEEE 802.3 10GE LAN MAC implementation" (’692 Patent, col. 9:64-67). This suggests the term is not generic but refers to a specific, standardized hardware block.
B. ’975 Patent, Claim 5
- The Term: "compares the protocol information in a first pass... and, if the first result is positive, compares the protocol information in a second pass"
- Context and Importance: This language describes the core logic of the invention. Infringement will likely depend on whether the accused FastLinQ adaptor performs this specific conditional, sequential comparison.
- Intrinsic Evidence for a Broader Interpretation: Practitioners may argue that the term covers any multi-step analysis where the result of an earlier step influences the handling of a later step, even if the steps are performed in parallel, pointing to general descriptions of a "multi-pass comparison" (’975 Patent, col. 4:11-12).
- Intrinsic Evidence for a Narrower Interpretation: The claim's "if... then" structure, along with the patent's flowcharts, strongly supports a narrow interpretation requiring a sequential and conditional process. The specification explains that if a match is not found in the first pass, "the process proceeds to block 560" where a default tag is assigned, bypassing the second comparison entirely (’975 Patent, col. 7:38-44; Fig. 5), reinforcing a hierarchical dependency.
VI. Other Allegations
A. Indirect Infringement
The complaint alleges induced infringement for all three patents, stating that Defendant knowingly provides its products to customers and end-users for "use in an infringing manner" (Compl. ¶¶ 23, 33, 43). The allegations do not point to specific instructions, user manuals, or other evidence of intent to induce.
B. Willful Infringement
Willfulness is alleged based on knowledge of the patents "at least as of the date of this Complaint" (Compl. ¶¶ 23, 33, 43). This allegation appears to be aimed at establishing post-filing willfulness, as no facts supporting pre-suit knowledge are presented.
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of evidentiary proof: The complaint makes conclusory allegations that the accused products perform the specific, multi-step functions recited in the claims. A key question for the court will be whether Plaintiff can produce technical evidence—beyond marketing materials—that demonstrates the accused Marvell products actually operate in the precise manner claimed, such as performing the '975 patent’s conditional two-pass comparison or using a specific "10GE LAN media access controller" for monitoring as required by the '692 patent.
- A second central question will be one of claim construction: The case may turn on the scope of key claim terms. For example, can the functionality of "VLAN tagging," as implemented in the accused FastLinQ adaptor, be construed to meet the detailed, hierarchical "protocol determination" logic required by claim 5 of the '975 patent, or is there a fundamental mismatch in technical operation?
- A final question relates to damages: Because the willfulness allegations are predicated on knowledge only as of the complaint's filing date, the viability of a claim for enhanced damages for any pre-suit infringement will depend on whether Plaintiff can uncover evidence of Defendant’s pre-suit knowledge of the patents.