DCT
6:20-cv-00473
WSOU Investments LLC v. Dell Tech Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: WSOU Investments, LLC d/b/a Brazos Licensing and Development (Delaware)
- Defendant: Dell Technologies Inc., Dell Inc., and EMC Corporation (Delaware/Massachusetts)
- Plaintiff’s Counsel: Etheridge Law Group, PLLC; Law Firm of Walt, Fair PLLC
- Case Identification: 6:20-cv-00473, W.D. Tex., 10/19/2020
- Venue Allegations: Venue is alleged to be proper based on Defendants having established places of business within the Western District of Texas, being registered to do business in Texas, and having transacted business and committed acts of infringement in the district.
- Core Dispute: Plaintiff alleges that Defendant’s Dell EMC SmartFabric Operating System infringes a patent related to methods for selecting communication paths to perform network load balancing.
- Technical Context: The technology concerns efficient data routing in complex computer networks, specifically methods for distributing traffic across multiple available equal-cost paths to maximize resource utilization and prevent bottlenecks.
- Key Procedural History: The complaint notes that Plaintiff filed a prior suit in May 2020 against the same Defendants asserting infringement of the same patent, which was subsequently dismissed. Plaintiff leverages this prior suit to allege that Defendants have had knowledge of the patent since at least May 2020. Subsequent to the filing of this complaint, an Ex Parte Reexamination Certificate for the patent-in-suit was issued on April 25, 2023, confirming the patentability of all 14 original claims.
Case Timeline
| Date | Event |
|---|---|
| 2012-09-28 | U.S. Patent No. 9,137,144 Priority Date |
| 2015-09-15 | U.S. Patent No. 9,137,144 Issue Date |
| 2020-05 | Plaintiff files prior suit against Defendants (approx. date) |
| 2020-10-19 | Complaint Filing Date |
| 2023-04-25 | U.S. Patent No. 9,137,144 Reexamination Certificate Issued |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 9,137,144 - "Method and Apparatus for Communication Path Selection," Issued September 15, 2015
The Invention Explained
- Problem Addressed: In computer networks with multiple routes between two points, standard protocols like Shortest Path Bridging (SPB) sometimes fail to use all available redundant paths for load balancing, leading to inefficient use of network resources (U.S. Patent No. 9,137,144, col. 2:5-14). Existing protocols may select only a limited subset of paths, leaving other perfectly viable, equal-cost paths unused.
- The Patented Solution: The patent discloses a method for distributing traffic groups (like Virtual LANs, or VLANs) across all available equal-cost, equal-hop (EHEC) paths. The method involves determining the total number of available lowest-cost paths ("N") between a source and destination, assigning a sequential identifier ("V") to a traffic group, and then using a modulo operation ("V mod N") to deterministically map that traffic group to one of the N paths via an indexed table (’144 Patent, col. 2:28-40; Fig. 4). This ensures traffic is spread across all available pathways.
- Technical Importance: The method aims to provide a more robust and complete utilization of network infrastructure than existing protocols by ensuring that all available low-cost paths are considered and used for traffic distribution (’144 Patent, col. 2:11-14).
Key Claims at a Glance
- The complaint asserts infringement of at least Claim 1 (Compl. ¶30).
- Independent Claim 1 requires:
- determining lowest cost paths from the plurality of contiguous communication paths;
- determining V mod N;
- comparing a result of the determining to indices on a path selection table that associates a unique index with each of the plurality of communication paths; and
- selecting a path associated with an index equal to the result, wherein N is a number of the lowest cost paths in the plurality of communication paths and V is a group identifier corresponding to the group of communication traffic.
- The complaint does not explicitly reserve the right to assert dependent claims but references infringement of "one or more claims" in the prayer for relief (Compl. p. 16, ¶A).
III. The Accused Instrumentality
Product Identification
- The Dell EMC SmartFabric Operating System (OS10) (Compl. ¶13).
Functionality and Market Context
- The complaint describes SmartFabric OS10 as a network operating system that provides Equal Cost Multi-Path (ECMP) routing, a technique for forwarding packets to a single destination over multiple "best paths" of equal cost (Compl. ¶19).
- To select a path, OS10 allegedly uses a "hash algorithm" that makes decisions based on values in packet fields (e.g., VLAN ID) and internal values (Compl. ¶15, ¶17). The complaint alleges this hashing involves a modulo operation, where a "hash key" is used with the number of available next-hops in an ECMP group to select an egress port (Compl. ¶21). A feature called "resilient hashing" is alleged to create a "flow-map table" with a default of 64 paths to equally distribute traffic (Compl. ¶25).
IV. Analysis of Infringement Allegations
’144 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| determining lowest cost paths from the plurality of contiguous communication paths | The Accused Products use ECMP to determine multiple "best paths," which are defined as the "lowest cost paths," i.e., paths with the minimum number of hops between source and destination. | ¶19, ¶20 | col. 9:11-13 |
| determining V mod N, | The Accused Products allegedly use a "resilient hashing" algorithm where the next hop is selected based on a "hash key modulo, the number of ports in a port channel or next-hops in an ECMP group." The complaint alleges the VLAN ID can be the "group identifier" (V) and the number of ECMP paths is "N." | ¶17, ¶21, ¶25 | col. 9:14 |
| comparing a result of the determining to indices on a path selection table that associates a unique index with each of the plurality of communication paths | The complaint alleges that when resilient hashing is enabled, a "flow-map table is created with 64 paths" and that this table contains "indices of the 64 lowest cost paths." The result of the hashing maps traffic to these paths. The complaint provides a diagram showing this mapping. This visual, titled "Traffic flow with resilient hashing enabled," depicts three distinct traffic flows being directed by a "Hashing" function to different next-hop channels. | ¶25, ¶26, ¶27 | col. 9:15-18 |
| selecting a path associated with an index equal to the result, wherein N is a number of the lowest cost paths... and V is a group identifier... | The Accused Products are alleged to use the result of the hashing to map traffic to a specific path/next-hop in the flow-map table. The complaint alleges the paths are "uniquely identified (based on a unique index) by to load balance the traffic." A visual titled "Member link goes down" illustrates how traffic is re-distributed when a path is deleted, implying a selection based on an indexed table. | ¶27, ¶28, ¶29 | col. 9:18-22 |
- Identified Points of Contention:
- Scope Questions: The case may turn on whether the accused "hash algorithm," which considers multiple packet fields, constitutes "determining V mod N," where "V" is specifically a "group identifier." A question for the court will be whether the patent’s "group identifier" can be construed to read on the complex, multi-factor input of a modern hashing algorithm, or if it implies a simpler, more direct mapping from a single identifier like a VLAN ID.
- Technical Questions: A key question is whether the accused "resilient hashing" function, which allegedly performs a modulo operation on the number of active links in a group, is technically equivalent to the claimed method of performing a modulo operation on "N," the total number of available lowest-cost paths. Further, it raises the question of whether the accused "flow-map table," a potentially dynamic structure, is the same as the "path selection table" described in the patent, which appears to be a more static, indexed list of all possible paths.
V. Key Claim Terms for Construction
The Term: "group identifier"
Context and Importance: This term is central to defining the input "V" in the "V mod N" calculation. Its construction will determine whether the claim covers sophisticated hashing algorithms or is limited to systems that use a single, designated identifier for a traffic group. Practitioners may focus on this term because the accused product's hash considers many packet fields, while the patent's examples focus heavily on VLAN IDs.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim itself uses the general term "group identifier". The specification also refers more broadly to "VLANs or other groups of data traffic" (’144 Patent, col. 2:21-22).
- Evidence for a Narrower Interpretation: The specification repeatedly uses VLANs as the primary, and often sole, example of a traffic group, stating that "the communication group is a VLAN" and "assigning a path to a VLAN" (’144 Patent, col. 2:38-39; Abstract).
The Term: "path selection table"
Context and Importance: The infringement theory hinges on equating the accused "flow-map table" with the claimed "path selection table". The defense will likely argue these are technically distinct. The construction will determine if the accused product's potentially dynamic, algorithmically generated table meets the limitation.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim defines the table by its function: one that "associates a unique index with each of the plurality of communication paths" (’144 Patent, Claim 1). Any structure performing this function could fall within its scope.
- Evidence for a Narrower Interpretation: The patent’s exemplary embodiments illustrate a specific structure: a well-ordered list where paths are stored and associated with a sequential index (e.g., Table 2, Table 4) (’144 Patent, col. 8). This could be argued to support a narrower construction limited to such static, indexed lists.
VI. Other Allegations
- Indirect Infringement: The complaint alleges induced infringement, stating Defendants provide "product descriptions, operating manuals, and other instructions on how to implement and configure the Accused Products" that encourage infringing use (Compl. ¶34). It also alleges contributory infringement, claiming the products are "especially made or adapted for infringing" and lack substantial non-infringing uses for the relevant features (Compl. ¶35).
- Willful Infringement: Willfulness is alleged based on Defendants having "notice and actual or constructive knowledge" of the patent since at least May 2020, due to a prior lawsuit filed by Plaintiff involving the same patent and products (Compl. ¶33).
VII. Analyst’s Conclusion: Key Questions for the Case
The resolution of this dispute will likely depend on the court's answers to two central questions:
A core issue will be one of definitional scope: Can the term "group identifier", as used in the patent, be construed to cover the multi-factor input of the accused product's general-purpose hashing algorithm, and is the accused "flow-map table" structurally and functionally the same as the patent’s claimed "path selection table"?
A key evidentiary question will be one of technical operation: Does the accused "resilient hashing" function, which allegedly calculates a hash modulo the number of active paths in a group, perform the same operation as the claimed method, which requires calculating "V mod N," where N is the total number of available lowest-cost paths between a source and destination?