Method of generating round robin service order lists for improving service performance during server failures

Patent metadata

Field	Value
Publication number	US-9241045-B2
Application number	US-201213407312-A
Country	US
Kind code	B2
Filing date	Feb 28, 2012
Priority date	Nov 7, 2008
Publication date	Jan 19, 2016
Grant date	Jan 19, 2016

Abstract

A method controls the routing of service requests to a plurality of servers using a first routing distribution algorithm. The method includes waiting a first period of time for a designated server to respond to a service request, transmitting the service request to the designated server a second time, and waiting a second period to time for the designated server to respond to the service request assigned to the designated server, the second period of time being longer than the first period of time. The method also includes determining that the designated server has failed, rerouting the service request to a different server, and routing the service requests to the plurality of servers using a second routing distribution algorithm.

First claim

What is claimed is:

1. A method for controlling the routing of service requests to a plurality of servers, the method comprising:
routing the service requests to the plurality of servers using a first routing distribution algorithm;
waiting a first period of time for a designated server of the plurality of servers to respond to a service request assigned to the designated server;
transmitting the service request to the designated server a second time when the designated server does not respond to the service request within the first period of time;
waiting a second period to time for the designated server to respond to the service request assigned to the designated server, the second period of time being longer than the first period of time;
determining that the designated server has failed when the designated server does not respond to the service request within the second period of time;
rerouting the service request assigned to the designated server that has failed to a different server of the plurality of servers;
routing the service requests to the plurality of servers using a second routing distribution algorithm,
wherein the routing and rerouting are performed by a border element in a voice over internet protocol system, and
wherein the first routing distribution algorithm is based on a round robin service order, and the second routing distribution algorithm is a modified version of the round robin service order of the first routing distribution algorithm that includes the designated server that has failed in a fixed order that does not vary, but varies the order of routing the service requests for all of the plurality of servers, except for the order of routing for the designated server that has failed, in a round robin service order.

2. The method described in claim 1 , further comprising:
waiting a third period of time for the designated server to respond to the service request assigned to the designated server, the third period of time being longer that the second period of time.

3. The method described in claim 1 ,
wherein a same percentage of service requests, of a total number of service requests routed to the plurality of servers, are assigned to the designated server that has failed after determining that the designated server has failed as before determining that the designated server has failed.

4. The method described in claim 1 ,
wherein the second routing distribution algorithm evenly distributes the service requests among the plurality of servers, including the designated server that has failed.

5. The method described in claim 1 , further comprising:
waiting a fourth period of time for the different server of the plurality of servers to respond to a service request assigned to the different server; and
transmitting the service request to the different server a second time when the different server does not respond to the service request within the fourth period of time.

6. The method described in claim 5 , further comprising:
waiting a fifth period of time for the different server to respond the service request assigned to the different server; and
blocking the service request when the different server does not respond to the service request within the fifth period of time.

7. A network communication node for controlling the routing of service requests to a plurality of servers, the network communication node comprising:
a request distribution manager that assigns the service requests to the plurality of servers using a first distribution algorithm; and
a router that routes assigned service requests to the plurality of servers over a communications network,
wherein the request distribution manager waits a first period of time for a designated server of the plurality of servers to respond to a service request assigned to the designated server;
wherein the request distribution manager has the router transmit the service request to the designated server a second time when the designated server does not respond to the service request within the first period of time;
wherein the request distribution manager waits a second period to time for the designated server to respond to the service request assigned to the designated server, the second period of time being longer than the first period of time;
wherein the request distribution manager determines that the designated server has failed when the designated server does not respond to the service request within the second period of time;
wherein the request distribution manager has the router reroute the service request assigned to the designated server that has failed to a different server of the plurality of servers;
wherein the request distribution manager assigns the service requests to the plurality of servers using a second routing distribution algorithm;
wherein the routing and rerouting are performed by a border element in a voice over internet protocol system, and
wherein the first routing distribution algorithm is based on a round robin service order, and the second routing distribution algorithm is a modified version of the round robin service order of the first routing distribution algorithm that includes the designated server that has failed in a fixed order that does not vary, but varies the order of routing the service requests for all of the plurality of servers, except for the order of routing for the designated server that has failed, in a round robin service order.

8. The network communication node described in claim 7 ,
wherein a same percentage of service requests, of a total number of service requests routed to the plurality of servers, are assigned to the designated server that has failed after determining that the designated server has failed as before determining that the designated server has failed.

9. A non-transitory computer-readable storage medium encoded with a computer program that, when executed, controls the routing of service requests to a plurality of servers over a communication network, comprising:
routing the service requests to the plurality of servers using a first routing distribution algorithm;
waiting a first period of time for a designated server of the plurality of servers to respond to a service request assigned to the designated server;
transmitting the service request to the designated server a second time when the designated service does not respond to the service request within the first period of time;
waiting a second period to time for the designated server to respond to the service request assigned to the designated server, the second period of time being longer than the first period of time;
determining that the designated server has failed when the designated server does not respond to the service request within the second period of time;
rerouting the service request assigned to the designated server that has failed to a different server of the plurality of servers;
routing the service requests to the plurality of servers using a second routing distribution algorithm,
wherein the routing and rerouting are performed by a border element in a voice over internet protocol system, and
wherein the first routing distribution algorithm is based on a round robin service order, and the second routing distribution algorithm is a modified version of the round robin service order of the first routing distribution algorithm that includes the designated server that has failed in a fixed order that does not vary, but varies the order of routing the service requests for all of the plurality of servers, except for the order of routing for the designated server that has failed, in a round robin service order.

Assignees

• Hoeflin David
• Bakshi Yury
• At & T Ip I Lp

Inventors

• Hoeflin David
• Bakshi Yury

Classifications

• H04L65/1063

  Application servers providing network services (systems providing special services to telephonic subscribers H04M3/42) · CPC title

• H04L69/40 Primary

  for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection (management of faults, events, alarms or notifications in data switching networks H04L41/06) · CPC title

• H04L67/327 Primary

  Electricity · mapped topic

• H04L67/63 Primary

  Routing a service request depending on the request content or context · CPC title