Google has received a patent for the technique known as MapReduce. The patent, number 7,650,331, applied for in 2004, is entitled "System and method for efficient large-scale data processing" and covers the process of mapping work to multiple processors and then reducing the intermediate results from these processors to a final result.
The technique is used widely by data mining companies, for example, in Yahoo's search infrastructure, Amazon's Elastic MapReduce service and IBM's M2 platform. The Apache Hadoop is the most prominent open source implementation of the technique.
The concept of mapping and reducing functions has been a fundamental idea behind distributed parallel processing for many years, and in a dispute it could be reasonably claimed that Google didn't invent MapReduce itself, but that would just move the argument on to the specific claims within the patent.
Google has told US media that "Like other responsible, innovative companies, Google files patent applications on a variety of technologies it develops. While we do not comment about the use of this, or any part of our portfolio, we feel that our behavior to date has been in line with our corporate values and priorities." Given that Google has not pursued patent infringements and appear to have been building a defensive patent portfolio, it is believed by some that Google is ensuring that it is not possible for a patent troll to obtain a similar patent and use it against Google and others.
[See contending views of MapReduce in Communications (Jan. 2010) MapReduce and Parallel DBMSs: Friends or Foes? by Michael Stonebraker et al and MapReduce: A Flexible Data Processing Tool by Jeffrey Dean and Sanjay Ghemawat and "MapReduce and Parallel DBMSs: Friends or Foes?" by Michael Stonebraker et al.]
From The H
View Full Article
No entries found