Scalable systems and methods for generating and serving recommendations

• US 10,497,044 B2
• Filed: 10/19/2015
• Issued: 12/03/2019
• Est. Priority Date: 10/19/2015
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Demandware, Inc. (Salesforce.com, Inc.)
• Inventors: Krishnan, Bharath K., Torra, Michael J.
• Primary Examiner(s): Misiaszek, Michael
• Assistant Examiner(s): Kang, Timothy J
• Application Number:
  US14/886,826
• Publication Number:
  US 20170109807A1
• Time in Patent Office: 1,506 Days
• Field of Search: None
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06Q 30/0253 During e-commerce, i.e. online transactions
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0261 based on user location
• CPC Class Codes: G06Q 30/0271 Personalized advertisement
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]

Abstract

A scalable recommendation engine includes stateless processors in communication with at least one memory server that stores contextual data. A router is configured to direct a recommendation request to a first stateless processor, which is configured to generate a recommendation using contextual data from the memory server. A controller monitors the available processing bandwidth of the stateless processors and deploys an additional stateless processor if the available processing bandwidth is less than a minimum available processing bandwidth threshold. The controller can remove from deployment a stateless processor if the available processing bandwidth is greater than a maximum pre-determined available processing bandwidth threshold.

Claims

1. A scalable recommendation engine comprising:

a plurality of stateless processors deployed without data needed to generate a recommendation;

at least one common memory server that is common to the plurality of stateless processors, the at least one common memory server being in communication with the plurality of stateless processors and storing contextual data;

a router in communication with the plurality of stateless processors to direct a recommendation request to a selected said stateless processor,

each of the plurality of stateless processors being configured to receive a said recommendation request from the router, to send a query for contextual data to the at least one common memory server, to receive that contextual data from the at least one common memory server, and to generate a recommendation using said contextual data; and

a controller in communication with the plurality of stateless processors, the controller configured to monitor an available processing bandwidth of said plurality of stateless processors

and to deploy an additional stateless processor without data needed to generate a recommendation, if the available processing bandwidth is less than a minimum available processing bandwidth threshold.

2. The scalable recommendation engine of claim 1, wherein the controller is configured to release an excess stateless processor from deployment if the available processing bandwidth is greater than a pre-determined maximum available processing bandwidth threshold.