System and method for efficient delivery in a multi-source, multi destination network

• US 8,086,692 B2
• Filed: 08/27/2008
• Issued: 12/27/2011
• Est. Priority Date: 08/27/2008
• Status: Expired due to Fees

• Current Assignee: Tech Mahindra Limited
• Original Assignee: Satyam Computer Services Limited (Tech Mahindra Limited)
• Inventors: Sridhar, Varadarajan, Gopalan, Srividya
• Primary Examiner(s): Nguyen, Phuoc
• Assistant Examiner(s): MURRAY, DANIEL C
• Application Number:
  US12/199,658
• Publication Number:
  US 20100057909A1
• Time in Patent Office: 1,217 Days
• Field of Search: 709/225, 709/217, 709/232, 707/802
• US Class Current: 709/217
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1017 based on a round robin mechanism

Abstract

Real-time delivery of multimedia content puts extreme demand on both computational and network resources. Many a times, the requirement is to handle multiple such delivery requirements and as a consequence, the overall delivery efficiency gets affected. Another dimension of complexity is that (a) the multimedia content to be delivered may not always be highly reliable, and (b) the network may also be not highly reliable. In such a situation, it becomes even more difficult to satisfy the quality of service requirements. When content to be delivered to multiple destinations is available at multiple sources, there is an opportunity to overcome the above mentioned complexity. The system and method for efficient delivery in a multi-source multi-destination network involves pulling of data from the multiple sources by measuring quality of service and reliability parameters, and efficiently loading the systems and the network in a cooperative manner.

Claims

1. A method for delivering a multimedia content through a network interconnecting a plurality of host systems, wherein said multimedia content is to be delivered from a plurality of source host systems of said plurality of host systems to a plurality of destination host systems of said plurality of host systems, wherein said multimedia content comprises a plurality of segments, wherein a segment of said plurality of segments is the amount of said multimedia content to be pulled from a source host system of said plurality of source host systems by a destination host system of said plurality of destination host systems, and said source host system delivers said segment in response to a pull request during a session with an overall priority to said destination host system, using a computer, said method comprising:

computing a normalized delay measure based on a quality of service measure associated with said source host system;

computing a normalized error measure based on said quality of service measure associated with said source host system;

computing a normalized loss measure based on a reliability measure associated with said source host system;

determining a normalized session priority measure based on a priority of a session related to said source host system;

computing a normalized transformed elapsed time measure based on an inverse normal distribution of a normalized elapsed time of said session;

computing a normalized user loading measure based on a load measure associated with said destination host system, said session, and a pre-defined maximum number of users;

computing a normalized session loading measure based on a load measure associated said source host system;

computing a normalized measure of a plurality of normalized measures based on said normalized delay measure, said normalized error measure, said normalized loss measure, said normalized session priority measure, said normalized transformed elapsed time measure, said normalized user loading measure, and said normalized session loading measure, wherein each of said plurality of normalized measures is associated with a host system of said plurality of source host systems;

computing a normalized weight related to said source host system based on said normalized measure associated with said source host system and said plurality of normalized measures;

computing said segment to be pulled from said source host system based on said multimedia content and said normalized weight related to said source host system;

determining a session priority associated with said session; and

computing said overall priority of said pull request based on said session priority, an elapsed time associated with said session, an amount of data associated with said pull request, and an accumulated amount of data associated with said pull request.