Obviousness

To analyze the obviousness of US Patent 9,800,633 under 35 U.S.C. § 103, we must consider the state of the art at the patent's priority date (July 18, 2012) and evaluate whether the claimed invention would have been obvious to a person having ordinary skill in the art (PHOSITA) by combining existing knowledge or prior art references.

The patent itself outlines the background art, which serves as a foundation for this obviousness analysis.

Background Art Acknowledged by the Patent

The patent's background and overview sections acknowledge several existing technologies and problems:

1. Mobile Video Streaming and Optimization: "Mobile devices, such as smart phones and tablets, have become prevalent in recent years... more and more users view streamed videos on their mobile devices." It also states that "Existing network providers or media sources have attempted to improve the bandwidth usage during video streaming by performing a video optimization on the streamed video." This establishes the knowledge of video optimization and transcoding for different device and network conditions.
2. Local Caching: "Network providers and media sources may use local caching as a method for improving the streaming quality of source files such as video to mobile devices." It further notes that "Local caching allows the video optimizer to store the transcoded source file (or optimized version of the source file) for a short duration and respond to a user's requests... without having to re-retrieve the source file for transcoding." This demonstrates that local caching of transcoded files at individual servers was known to reduce delays for repeat requests within a single session.
3. Problem with Redundant Transcoding in Distributed Systems: The patent explicitly identifies a key problem: "Even if cached files are not purged, a user has to be fortunate enough to be routed to a particular video optimizer, out of potentially thousands, having cached the optimized source file. Otherwise, the user has to wait through the same transcoding operation as the user who had previously streamed the video from another video optimizer." This highlights the inefficiency and redundancy when multiple independent video optimizers exist, but lack a mechanism to share their cached, optimized content effectively.
4. Components of an ID Key: The patent mentions that a "source file signature may comprise a hash of the original file or portion of the original file, attributes of the video (e.g., bit rate, frame width, frame height, sample rate, audio properties and the like), URL of the original file and so forth." This indicates that methods for uniquely identifying original files and specifying optimization parameters were individually known.

Obviousness Analysis of Independent Claims (Claims 1, 9, and 16)

The independent claims (Claim 1, 9, and 16) describe a system and method for a "first video optimizer" to receive a request with a specific "ID key" (identifying both the original source file and the desired optimization parameters), query a "cache database" to find if a "second video optimizer" has the matching optimized file, and then retrieve and stream that file to the user, thereby avoiding redundant transcoding.

A PHOSITA in the field of video streaming, content delivery networks (CDNs), and network optimization at the priority date would possess knowledge of the following concepts:

• Video transcoding and optimization: Adapting video content to various device capabilities and network conditions.
• Distributed computing and networks: Systems comprising multiple servers (like video optimizers) cooperating to provide a service.
• Load balancing and content redirection: Directing user requests to appropriate servers, often based on server load or content availability.
• Centralized databases or registries: Used in distributed systems to track resources, content, or server locations (e.g., for CDNs, distributed file systems, or service discovery).
• Content identification: Using hashes, URLs, or metadata to uniquely identify content.

Obviousness Combination:
A PHOSITA, when presented with the problem of "redundant transcoding" across "potentially thousands" of video optimizers that individually use "local caching", would have a strong motivation to combine existing technologies to address this inefficiency. The obvious solution would be to implement a mechanism for these distributed optimizers to share information about their cached content.

Proposed Combination of Prior Art for Obviousness:

1. Distributed Video Optimizers with Local Caches: As acknowledged by the patent, individual video optimizers performing optimization and maintaining "local caching" of transcoded files were known. This forms the basis of the "first video optimizer" and "second video optimizer" with their "local caches."
2. Centralized Resource Registry/Database in a Distributed System: It would be obvious for a PHOSITA managing a distributed system (like a cluster of video optimizers) to implement a "cache database" (Prior Art 4) or registry to keep track of resources available across the network. This is a common architectural pattern in distributed computing and content delivery networks to improve efficiency and avoid duplication of effort.
3. Unique Content Identification with Parameters: To make the centralized database useful for video optimization, the cached entries need to be uniquely identifiable. It would be obvious for a PHOSITA to use an ID key that not only identifies the "original source file" (e.g., via hash or URL as described in the patent) but also incorporates the "optimization parameters" (e.g., bit rate, frame size, etc. as described in the patent). This is because the 'optimized version' of a video is defined by these parameters, and a client will only benefit from a cached version if it matches the required parameters. The combination of these known identification elements into a single lookup key for a distributed cache is a straightforward design choice.
4. Inter-Server Communication and Redirection: Once the central database identifies an optimizer holding the desired optimized file, it would be obvious to either retrieve the file from that optimizer (server-to-server transfer, a standard CDN function) or redirect the user device directly to that optimizer (HTTP redirection, a standard load balancing and CDN function).

Motivation for Combination:

The motivation for a PHOSITA to combine these elements is directly articulated in the patent's background: to overcome the inefficiency and delays caused by redundant transcoding in a distributed video optimization environment where individual local caches do not communicate. By implementing a central "cache database" to track which "optimized version" of a "source file" (identified by a combined ID key) is stored in which "local cache" of a "video optimizer," the system can avoid re-transcoding, thus reducing processing load, saving bandwidth, and improving user experience by providing optimized content "just-in-time" from an already existing source. This directly addresses the problem of users having to "wait through the same transcoding operation as the user who had previously streamed the video from another video optimizer" due to a lack of shared knowledge across optimizers.

Therefore, the core concept of using a centralized database to coordinate local caches of optimized video files across multiple optimizers, identified by an ID key representing both the original file and its optimization parameters, would have been obvious to a PHOSITA seeking to improve efficiency in distributed video streaming systems.