Reliable intermediate multicast communications

Patent metadata

Field	Value
Publication number	US-9753797-B1
Application number	US-201113218768-A
Country	US
Kind code	B1
Filing date	Aug 26, 2011
Priority date	Aug 26, 2011
Publication date	Sep 5, 2017
Grant date	Sep 5, 2017

Abstract

In certain embodiments, a computer-implemented method includes detecting, by a first processing system, that a failure event operable to cause the first processing system to become inoperable has occurred. The method further includes communicating, by the first processing system using multicasting, a notification to two or more proxy processing systems. At least one of the proxy processing systems is operable to receive the notification and communicate the notification to a second processing system on behalf of the first processing system using a reliable communications protocol.

First claim

What is claimed is:

1. A system, comprising:
a first processing system;
a plurality of proxy processing systems separate from the first processing system; and
a server, wherein:
the server is part of a network of processing systems of which the first processing system and the plurality of proxy processing systems are a part;
in response to a failure event of the first processing system detected by a network interface of the first processing system, the network interface generates a notification to the plurality of proxy processing systems, the notification comprising an indication of the failure event and communicated on behalf of the first processing system to each proxy processing system of the plurality of proxy processing systems using multicasting, the failure event causing the first processing system to be at least partially inoperable by a software or hardware failure; and
each proxy processing system of the plurality of proxy processing systems at least:
receives the notification;
determines that a different proxy processing system of the plurality of proxy processing systems also received the notification as a result of the multicasting;
detects a parameter associated with the plurality of proxy processing systems;
communicates with the different proxy processing system to determine, based at least in part on the detected parameter, a selected proxy processing system of the proxy processing system and the different proxy processing system; and
causes the selected proxy processing system to communicate, using a reliable communications protocol, the notification to the server on behalf of the first processing system.

2. The system of claim 1 , wherein the network interface of the first processing system is further operable to:
determine a timestamp indicating when the failure event occurred;
access data associated with the failure event; and
formulate the notification using at least the timestamp and the data prior to communicating the notification.

3. The system of claim 1 , wherein the network interface of the first processing system is further operable to:
access a plurality of pre-crafted notification messages stored in one or more memory devices coupled to the first processing system;
compare the plurality of pre-crafted notification messages with the failure event; and
select, based at least in part on the comparison, a pre-crafted notification message from the plurality of pre-crafted notification messages to include in the notification communicated to the plurality of proxy processing systems.

4. The system of claim 1 , wherein the failure event comprises one or more of the following errors:
a power failure;
a software crash;
a hardware malfunction; or
a temperature of the first processing system exceeding a predetermined temperature.

5. The system of claim 1 , wherein the server is operable to receive the notification, and in response, cause traffic to be routed away from the first processing system.

6. The system of claim 1 , wherein at least one of the plurality of proxy processing systems is integrated with the first processing system.

7. The system of claim 1 , wherein the first processing system and the second processing system are servers.

8. A computer-implemented method, comprising:
detecting, by a first processing system, that a failure event has occurred, the failure event operable to cause the first processing system to become at least partially inoperable due to a software or hardware failure; and
communicating, through a network interface of the first processing system including circuitry that detects the failure event, a multicasted notification to a plurality of proxy processing systems separate from the first processing system, the notification comprising an indication of the failure event, to cause at least one proxy processing system of the plurality of processing systems to at least:
determine that a different proxy processing system of the plurality of proxy processing systems also received the notification;
detect a parameter associated with the plurality of proxy processing systems;
communicate with the different proxy processing system to determine, based at least in part on the detected parameter, a selected proxy processing system of the proxy processing system and the different proxy processing system; and
cause the selected proxy processing system to communicate the notification to a second processing system on behalf of the first processing system using a reliable communications protocol.

9. The computer-implemented method of claim 8 , further comprising:
determining, by the network interface, a timestamp indicating when the failure event occurred;
accessing, by the network interface, data associated with the failure event; and
formulating, by the network interface prior to communicating the notification, the notification using at least the timestamp and the data.

10. The computer-implemented method of claim 8 , further comprising selecting a pre-crafted notification message from a plurality of pre-crafted notification messages to include in the notification communicated to the two or more proxy processing systems, wherein the plurality of pre-crafted notification messages are stored in one or more memory devices coupled to the first processing system.

11. The computer-implemented method of claim 8 , wherein the failure event comprises one or more of the following errors:
a power failure;
a software crash;
a hardware malfunction; or
a temperature of the first processing system exceeding a predetermined temperature.

12. The computer-implemented method of claim 8 , further comprising receiving the notification at the second processing system, and in response, causing traffic to be routed away from the first processing system.

13. The computer-implemented method of claim 8 , further comprising transmitting, by the at least one of the plurality of proxy processing systems, an indication that causes at least one other proxy processing system of the plurality of proxy processing systems not to communicate the notification to the second processing system.

14. A non-transitory computer-readable medium having stored thereon one or more executable instructions that, when executed by one or more processing units of a computer system, cause the computer system to at least:
receive a multicasted notification from a first processing system separate from the computer system, communication of the notification initiated by a network interface of the first processing system as a result of the first processing system becoming at least partially inoperable, the notification multicasted by the network interface to each of a plurality of proxy processing systems that includes the computer system, the notification comprising an indication of the failure event;
determine that a different proxy processing system of the plurality of proxy processing systems also received the notification;
detect a parameter associated with the plurality of proxy processing systems;
communicate with the different proxy processing system to determine, based at least in part on the detected parameter, a selected proxy processing system of the proxy processing system and the different proxy processing system; and
cause the selected proxy processing system to communicate the notification to a second processing system on behalf of the first processing system, the notification communicated to the second processing system using a reliable communications protocol.

15. The non-transitory computer-readable medium of claim 14 , wherein the notification comprises a User Da

Assignees

• Higgins Chris
• Amazon Tech Inc

Inventors

• Higgins Chris

Classifications

• G06F11/0772

  Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers · CPC title

• G06F11/0784 Primary

  Routing of error reports, e.g. with a specific transmission path or data flow · CPC title

• G06F11/0748 Primary

  in a remote unit communicating with a single-box computer node experiencing an error/fault (remote testing G06F11/2294) · CPC title

• H04L41/0668

  by dynamic selection of recovery network elements, e.g. replacement by the most appropriate element after failure · CPC title