PTAB

IPR2014-01399

Riverbed Technology Inc v. Parallel Networks LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Method and System for Uniform Resource Locator Transformation
  • Brief Description: The ’217 patent relates to a web caching system that transforms a requested Uniform Resource Locator (URL). The technology addresses the problem of caching content that varies based on HTTP request header information, not just the URL itself, by modifying the original URL to incorporate that header information. This allows different versions of the same content to be uniquely identified and cached using the new, transformed URLs.

3. Grounds for Unpatentability

Ground 1: Claims 1-20 are obvious over Hu in view of Mattis

  • Prior Art Relied Upon: Hu (Patent 6,173,322) and Mattis (Patent 6,292,880).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Hu taught a network request manager that functions as a cache server and uses pre-defined static rules to transform client requests. These transformations could be based on information parsed from the request, including header data like browser type, to modify a URL (e.g., by appending a new string). Petitioner contended that while Hu disclosed transforming a URL and caching content, it did not explicitly teach using the newly transformed URL to retrieve the corresponding content from its data cache. Mattis allegedly supplied this missing element. Mattis described a proxy cache for storing multiple versions ("alternates") of content that correspond to a single URL. Critically, Mattis taught creating a "name key" (a transformed URL) by concatenating the original URL with header information (e.g., the User-Agent header) and using this name key to cache and retrieve the specific content alternate.
    • Motivation to Combine: Petitioner asserted that a person of ordinary skill in the art (POSITA) seeking to implement Hu's caching system for header-dependent content would have been motivated to look to conventional caching techniques. This search would have led to Mattis, which provided a well-known method for caching content variants using modified URLs. A POSITA would combine Hu's URL transformation rules with Mattis's method of caching based on a transformed URL, viewing it as a simple substitution of one known caching technique for another to achieve the predictable result of correctly storing and retrieving different content versions.
    • Expectation of Success: A POSITA would have a high expectation of success in combining the references, as both described established techniques in the same field of web caching to yield a predictable improvement.

Ground 2: Claims 1-20 are obvious over Rosenberg in view of Apache

  • Prior Art Relied Upon: Rosenberg (Patent 5,740,430) and Apache (a technical publication titled "URL Manipulation with Apache").
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Rosenberg disclosed the core concept of caching dynamically generated, customized web pages. Rosenberg taught transforming a standard URL into a "Customized Uniform Resource Locator" (CURL) by combining the original URL with a selected attribute from the client request, such as the browser type found in the header. This new CURL was then used to find a matching customized file in the cache. However, Rosenberg did not specify the implementation details of how this transformation was performed. Petitioner argued that Apache provided these missing details. The Apache document described its mod_rewrite module, a well-known tool that uses rules with regular expressions to flexibly manipulate URLs. Apache explicitly showed how these rules could rewrite a URL based on comparing strings in both the URL and the HTTP header (e.g., the User-Agent field).
    • Motivation to Combine: A POSITA implementing Rosenberg's system would need a concrete method to perform the URL-to-CURL transformation. Petitioner argued that it would have been obvious to use Apache's mod_rewrite module, a widely known and available tool designed for exactly this purpose. Rosenberg itself acknowledged Apache as a known web server, making the use of its URL rewriting module a natural and logical choice to supply the missing implementation details for Rosenberg's conceptual framework.
    • Expectation of Success: The combination would have yielded predictable results because mod_rewrite was a robust and established tool for URL transformation. Applying its specific rules to implement Rosenberg's more general concept of a CURL would have been a straightforward engineering task with a high likelihood of success.

4. Key Claim Construction Positions

  • Petitioner argued that the claim term establish[ing] at a cache server a first uniform resource identifier and a header portion associated with a first content item was not clearly defined in the specification.
  • Under the broadest reasonable interpretation, Petitioner proposed construing this term to mean "associating a specific transform with a first uniform resource identifier and a header portion associated with the first content item." This construction was central to their invalidity arguments, as it allowed the pre-defined rewriting rules and regular expressions described in the prior art (the "specific transform") to be mapped directly onto the claim language.

5. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-20 of Patent 7,571,217 as unpatentable.