US 7,734,726 B2Granted

System and method for dynamically allocating processing on a network amongst multiple network servers

AssigneeInternational Business Machines Corporation

InventorsPaul Michael Dantzig, Daniel Manuel Dias, Richard Pervin King

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present invention provides a method, apparatus, and computer implemented instructions for processing Web and other Internet or Intranet based services. The system for processing Web requests includes a Web server with a connection to the Internet or Intranet with a predefined network bandwidth, a set of primary Web and application server cluster nodes to process the requests, and a dispatcher to allocate requests to nodes; in addition, one or more offload server nodes are connected to the network. Client Web requests arrive at the dispatcher of the Web server, which determines whether the incoming request can be handled at the primary Web server cluster, whether all or part of the user Web request should be offloaded to one of the offload server nodes, or whether the request should be throttled. If the dispatcher determines that the request should be handled by the primary Web server cluster, it is appropriately routed to one of the nodes in the primary Web server cluster; else if the dispatcher determines that the request should be offloaded, one of the offload server nodes or service providers is selected, and the request is either routed to a primary server node with the appropriate indication to offload all or part of the request, or the request is routed to the selected offload service provider; otherwise, the request is throttled by either routing it to a node which returns information that the service is overloaded, or if the Web servers are too busy to provide even an overload indication, then the request is dropped.

Claims16

Dependent claims

A method, in a network comprising a primary server and a plurality of offload servers, for dynamic offloading of processing requests from said primary server to any one of said plurality of offload servers, the method comprising steps of:

determining a load on said primary server;
if the load on said primary server is less than a first threshold, serving processing requests at said primary server;
only if the load on said primary server exceeds said first threshold, then offloading at least a portion of said processing requests to any one of said plurality of offload servers while said primary server continues to serve a remainder of said processing requests, wherein said any one of said plurality of offload servers processes said at least a portion of said processing requests and is associated with an offload threshold and the at least a portion of said processing requests is the only work handled by said any one of said plurality of offload servers, wherein the offloading is performed in accordance with a respective offload threshold for each of the plurality of offload servers, such that if said respective offload threshold is exceeded for every one of the plurality of offload servers, said offloading is stopped until a load on one of said plurality of offload servers falls below said respective offload threshold;
and
if the load on said primary server exceeds a second threshold, throttling at least one of said processing requests,
wherein serving the processing requests at said primary server includes returning a page to a user wherein all embedded objects in the page have links to said primary server;
and
wherein offloading at least a portion of the processing requests to any one of said plurality of offload servers includes serving a base page at said primary server in which links for embedded objects point to any one of said plurality of offload servers.

5 more elements

A method for allocating processing requirements on an Internet Protocol network between a primary server and a plurality of offload servers, comprising:

periodically evaluating processing requests to determine a load on said primary server;
if said load exceeds a first threshold, for a predetermined period of time directing at least one of said processing requests to any one of said plurality of offload servers while said primary server continues to serve a remainder of said processing requests, wherein said any one of said plurality of offload servers processes said at least one of said processing requests and is associated with an offload threshold and said at least one of said processing requests is the only work handled by said any one of said plurality of offload servers, wherein the directing is performed in accordance with a respective offload threshold for each of the plurality of offload servers, such that if said respective offload threshold is exceeded for every one of the plurality of offload servers, said offloading is stopped until a load on one of said plurality of offload servers falls below said respective offload threshold;
only if said load does not exceed said first threshold, directing said processing requests to said primary server;
and
if the load on said primary server exceeds a second threshold, throttling at least one of said processing requests, wherein directing said processing requests to said primary server further includes returning a page to a user wherein all embedded objects in the page have links to said primary server;
and
directing at least one processing request to any one of said plurality of offload servers further includes serving a base page at said primary server in which links for embedded objects point to said any one of said plurality of offload servers.

4 more elements

Dependencies are inferred from the claim text.

Facts

Application: 09/995,087Filed 27 Nov 2001
Published: US 2003/0101265 A129 May 2003
Granted: 8 Jun 2010
Term ends: Not estimatedThe source records don’t support an estimate.

Classification8 codes

G06F 9/505
The resource being a machine, e.g. CPUs, Servers, Terminals: considering the load
H04L 67/1008
Server selection for load balancing: based on parameters of servers, e.g. available memory or workload
H04L 67/1012
Server selection for load balancing: based on compliance of requirements or conditions with available server resources
G06F 2209/509
Indexing scheme relating to G06F9/50: offload