Content delivery network balancer

Patent metadata

Field	Value
Publication number	US-10440148-B1
Application number	US-201816030341-A
Country	US
Kind code	B1
Filing date	Jul 9, 2018
Priority date	Oct 30, 2015
Publication date	Oct 8, 2019
Grant date	Oct 8, 2019

Abstract

Techniques are described for balancing traffic among content delivery networks (CDNs). A CDN balancer can analyze CDN performance data and determine CDN balancing ratios used to distribute traffic among the CDNs.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
determining a first balance represented by a first plurality of weights or ratios for distributing content delivery traffic among a plurality of content delivery networks (CDNs) based on first performance data representing operation of the CDNs;
providing first manifest data to first client devices, the first manifest data being configured to enable the first client devices to access first media content via the CDNs in accordance with a first weight or ratio in the first plurality of weights or ratios;
providing second manifest data to second client devices, the second manifest data being configured to enable the second client devices to access the first media content via the CDNs in accordance with a second weight or ratio in the first plurality of weights or ratios;
determining a second balance represented by a second plurality of weights or ratios for distributing content delivery traffic among the CDNs based on second performance data representing operation of the CDNs, wherein the first performance data represent operation of the CDNs during a first time period and the second performance data represent operation of the CDNs during a second time period that is later than the first time period; and
providing third manifest data to third client devices, the third manifest data being configured to enable the third client devices to access second media content via the CDNs in accordance with a first weight or ratio in the second plurality of weights or ratios; and
providing fourth manifest data to fourth client devices, the fourth manifest data being configured to enable the fourth client devices to access the second media content via the CDNs in accordance with a second weight or ratio in the second plurality of weights or ratios.

2. The method of claim 1 , wherein the second time period is overlapping with the first time period.

3. The method of claim 1 , wherein determining the second balance is responsive to elapse of a period of time since determining the first balance, or wherein determining the second balance is responsive to occurrence of a first number of client device connections to one or more of the CDNs since determining the first balance.

4. The method of claim 1 , wherein the first and second performance data represent one or more of (1) rebuffering events for streaming sessions associated with the CDNs, (2) fatal errors for streaming sessions associated with the CDNs, (3) bitrate information for streaming sessions associated with the CDNs, (4) number of streaming sessions for each of the CDNs, or (5) capacity information for each of the CDNs.

5. The method of claim 1 , wherein the first and third client devices are represented by a single set of device characteristics, and wherein the first and second performance data only represent streaming sessions involving client devices represented by the single set of device characteristics.

6. The method of claim 1 , wherein the first client devices are represented by a first set of device characteristics and the third client devices are represented by a second set of device characteristics that is different from the first set of device characteristics, wherein the first performance data only represent streaming sessions involving client devices represented by the first set of device characteristics, and wherein the second performance data only represent streaming sessions involving client devices represented by the second set of device characteristics.

7. The method of claim 1 , wherein the first manifest data represent a first priority ordering of the CDNs, and wherein the second manifest data represent a second priority ordering of the CDNs.

8. The method of claim 1 , wherein the first and second performance data correspond to a rolling time period, the rolling time period including a less recent portion and a more recent portion, and wherein the first and second performance data corresponding to the more recent portion of the rolling time period are weighted more than the first and second performance data corresponding to the less recent portion of the rolling time period in determining the first and second balances.

9. The method of claim 1 , wherein the second balance of the content delivery traffic results in a traffic shift among the CDNs relative to the first balance of the content delivery traffic, and wherein the traffic shift is limited by a maximum threshold traffic shift, or wherein the traffic shift is at least as great as a minimum threshold traffic shift.

10. The method of claim 1 , wherein the second balance of the content delivery traffic results in a traffic increase for a first CDN relative to the first balance of the content delivery traffic, the traffic increase to the first CDN being based on capacity information for the first CDN.

11. A system, comprising one or more computing devices configured to:
determine a first balance represented by a first plurality of weights or ratios for distributing content delivery traffic among a plurality of content delivery networks (CDNs) based on first performance data representing operation of the CDNs;
provide first manifest data to first client devices, the first manifest data being configured to enable the first client devices to access first media content via the CDNs in accordance with a first weight or ratio in the first plurality of weights or ratios;
provide second manifest data to second client devices, the second manifest data being configured to enable the second client devices to access the first media content via the CDNs in accordance with a second weight or ratio in the first plurality of weights or ratios;
determine a second balance represented by a second plurality of weights or ratios for distributing content delivery traffic among the CDNs based on second performance data representing operation of the CDNs, wherein the first performance data represent operation of the CDNs during a first time period and the second performance data represent operation of the CDNs during a second time period that is later than the first time period; and
provide third manifest data to third client devices, the third manifest data being configured to enable the third client devices to access second media content via the CDNs in accordance with a first weight or ratio in the second plurality of weights or ratios; and
provide fourth manifest data to fourth client devices, the fourth manifest data being configured to enable the fourth client devices to access the second media content via the CDNs in accordance with a second weight or ratio in the second plurality of weights or ratios.

12. The system of claim 11 , wherein the second time period is overlapping with the first time period.

13. The system of claim 11 , wherein the one or more computing devices are configured to determine the second balance responsive to elapse of a period of time since determining the first balance, or responsive to occurrence of a first number of client device connections to one or more of the CDNs since determining the first balance.

14. The system of claim 11 , wherein the first and second performance data represent one or more of (1) rebuffering events for streaming sessions associated with the CDNs, (2) fatal errors for streaming sessions associated with the CDNs, (3) bitrate information for streaming sessions associated with the CDNs, (4) number of streaming sessions for each of the CDNs, or (5) capacity information for each of the CDNs.

15. The system of claim 11 , wherein the first and third client devices are represented by a single set of device characteris

Assignees

• Amazon Tech Inc

Inventors

• Richter Stefan Christian
• Benson Nicholas James
• Binns Justin Michael
• Burford Kenneth Thomas
• Freeman Ii James Marvin
• Joliveau Marc
• Tong Li

Classifications

• H04L43/0876

  Network utilisation, e.g. volume of load or congestion level · CPC title

• H04L43/0894

  Packet rate · CPC title

• H04L67/1008

  based on parameters of servers, e.g. available memory or workload (monitoring of computer activity G06F11/30) · CPC title

• H04L67/16

  Electricity · mapped topic

• H04L47/25

  with rate being modified by the source upon detecting a change of network conditions · CPC title