Dish Network LLC v. Sound View Innovations LLC

Prosecution History analysis

Examiner Search analysis

Invalidity Framework analysis

Petition

Petition Analysis

## 1. Case Identification
- Case #: [IPR2020-01041](https://ai-lab.exparte.com/case/ptab/IPR2020-01041)
- Patent #: [6,725,456](https://ai-lab.exparte.com/patent/6725456)
- Filed: June 5, 2020
- Petitioner(s): [DISH Network L.L.C.](https://ai-lab.exparte.com/party/dish-network-corp), DISH Technologies L.L.C., and [Sling TV L.L.C.](https://ai-lab.exparte.com/party/sling-tv-llc)
- Patent Owner(s): [Sound View Innovations, LLC](https://ai-lab.exparte.com/party/sound-view-innovations-llc)
- Challenged Claims: 13

## 2. Patent Overview
- Title: Quality of Service Method for Computer Systems
- Brief Description: The [’456](https://ai-lab.exparte.com/case/ptab/IPR2020-01041/doc/1001) patent relates to techniques for providing a desired quality of service (QoS) for an application running in a computer system. The disclosed method uses an operating system's application programming interface (API) to establish QoS guarantees via hierarchically organized resource reservations.

## 3. Grounds for Unpatentability

#### Ground 1: Obviousness over Durand in view of Bennett - Claim 13 is obvious over Durand in view of Bennett.
  - **Prior Art Relied Upon:** [Durand](https://ai-lab.exparte.com/case/ptab/IPR2020-01041/doc/1006) (Patent 6,338,072) and [Bennett](https://ai-lab.exparte.com/case/ptab/IPR2020-01041/doc/1008) (a 1997 article titled “Hierarchical Packet Fair Queueing Algorithms”).
  - **Core Argument for this Ground:**
    - **Prior Art Mapping:** Petitioner argued that Durand taught a method for ensuring QoS in a UNIX-based system by dynamically controlling resource allocation. Durand's system used a Local Resource Manager (LRM) API to organize processes into groups called "dimensions" and allocated resources based on relative weights assigned to these dimensions. Petitioner asserted this disclosed the core method of using an API to establish QoS guarantees and provide service to a request (limitations 13[a], 13[b], and 13[c]). However, Durand's "dimensions" were organized as peers, not hierarchically. Petitioner contended that Bennett remedied this deficiency by disclosing an idealized Hierarchical Generalized Processor Sharing (H-GPS) model. Bennett described a tree-like structure of nodes (resource reservations) where each node can have a parent, children, and siblings, and resources are distributed from a parent to its children based on proportional "service shares" (weights). This, Petitioner argued, supplied the hierarchical organization, parent-child relationships, and weight-based resource sharing recited in limitations 13[d] and 13[e].
    - **Motivation to Combine:** A POSITA would have been motivated to combine the references to solve the common problem of resource allocation. Petitioner argued a POSITA would improve Durand's "flat" dimension-based system by incorporating Bennett's well-known and superior hierarchical scheduling. This combination would provide more granular and predictable control over resource allocation among multiple jobs within a given dimension, which was a known technique to improve such systems.
    - **Expectation of Success:** A POSITA would have had a high expectation of success because the concepts in Durand and Bennett are compatible. Migrating Durand's relative weight approach to Bennett's more structured H-GPS hierarchy was a predictable implementation that would yield the expected benefits of enhanced resource control without undue experimentation.

#### Ground 2: Obviousness over Goyal in view of Bennett - Claim 13 is obvious over Goyal in view of Bennett.
  - **Prior Art Relied Upon:** [Goyal](https://ai-lab.exparte.com/case/ptab/IPR2020-01041/doc/1007) (a 1996 paper titled “A Hierarchical CPU Scheduler for Multimedia Operating Systems”) and Bennett (the same 1997 article).
  - **Core Argument for this Ground:**
    - **Prior Art Mapping:** Petitioner asserted that Goyal, unlike Durand, already disclosed a hierarchical resource allocation framework. Goyal taught a hierarchical CPU scheduler for a UNIX-derivative OS that partitioned CPU bandwidth among application classes using a tree structure. This system used a QoS manager and an API (system calls) to create nodes in the tree, assign weights, and provide QoS guarantees, thus teaching the substance of limitations 13[a] through 13[d]. Each node in Goyal’s tree has a name like a UNIX filename, which serves as a reference to an object. Petitioner contended that Bennett was best combined with Goyal to explicitly teach the final limitation (13[e]). While Goyal described allocating resources based on weights, Bennett provided explicit examples and clear rules for how a "guaranteed service rate" for a parent node constrains the sum of the guaranteed rates for its child nodes, ensuring the sum does not exceed the parent's allocation.
    - **Motivation to Combine:** Petitioner argued a POSITA would be motivated to combine the references because both addressed hierarchical resource allocation. Goyal taught that providing bounds on minimum throughput was desirable. Bennett explicitly solved this problem by introducing a "guaranteed service rate" for each node in its service hierarchy. A POSITA reading Goyal would have been motivated to look to a reference like Bennett to implement a more robust mechanism for guaranteeing minimum resource levels, thereby improving Goyal’s system in a predictable way.
    - **Expectation of Success:** Success would be highly predictable because both references describe compatible hierarchical resource allocation schemes. Integrating Bennett’s specific "guaranteed service rate" mechanism into Goyal’s existing hierarchical framework would be a straightforward enhancement for a POSITA, as it involves applying a known technique (guaranteed rates) to a similar system (Goyal's scheduler) to achieve a predictable result (improved QoS).

## 4. Key Claim Construction Positions
- **“correspond to a reference to an object”**: Petitioner argued this term should be construed to include at least indirect references, such as file descriptors or pointers, to various computing objects like CPU processes, memory objects, disk or network queues, and files. This construction is based on disclosures in the ’456 patent and was presented as important for mapping the prior art's use of process identifiers and filenames to the claim language.

## 5. Arguments Regarding Discretionary Denial
- Petitioner argued that the Board should not exercise discretionary denial under §314(a) or §325(d). It contended that this petition is not cumulative to prior IPRs filed by other parties (Walmart, Cigna) because it relies on different prior art references (Durand) and presents different theories of obviousness. Furthermore, Petitioner argued the key references (Durand and Goyal) were not substantively evaluated by the examiner during the original prosecution. Durand was never cited, and while Goyal was listed in the ’456 patent, the examiner never used it as a basis for rejection or explained why the claims were allowable over it.

## 6. Relief Requested
- Petitioner requested that the Board institute an *inter partes* review and cancel claim 13 of the ’456 patent as unpatentable.

DISH Network targets a Quality of Service patent on two obviousness grounds: Durand & Goyal cited against claim 13's hierarchical resource reservation.