Methods and computer program products for managing application performance on a network

• US 8,005,935 B2
• Filed: 04/05/2007
• Issued: 08/23/2011
• Est. Priority Date: 04/05/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Singh, Manpreet, Sahu, Sambit, Saha, Debanjan, Pradhan, Prashant
• Primary Examiner(s): SHAW, PELING ANDREW
• Application Number:
  US11/696,975
• Publication Number:
  US 20080250135A1
• Time in Patent Office: 1,601 Days
• Field of Search: 709/223, 709/224, 709/226, 709232-235, 715/734, 715/736, 370229-232, 370/338, 370/351, 370/468
• US Class Current: 709/223
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/14 Network analysis or design
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/0882 Utilisation of link capacity

Abstract

Managing application performance on a network. A network graph is generated from a set of application endpoints on the network. The network graph is annotated by associating one or more of the application endpoints with at least one of a corresponding latency annotation or a corresponding available bandwidth annotation. An end to end plan is formulated from the annotated network graph, wherein the plan maps one or more application requirements onto the plurality of paths in accordance with one or more end to end control adjustments including at least one of a server selection adjustment, an internet service provider (ISP) selector, or an overlay routing adjustment.

Claims

1. A method for managing application performance on a network, the method including:

generating a network graph from a set of application endpoints on the network, the network graph being indicative of a plurality of paths between the application endpoints, the plurality of paths each including a plurality of available resources on the network;

annotating the network graph by associating one or more of the application endpoints with a corresponding latency annotation and a corresponding available bandwidth annotation, wherein the latency annotation includes one or more application performance constraints and the available bandwidth notation includes one or more capacity constraints; and

formulating an end to end plan from the annotated network graph wherein the end to end plan maps one or more application requirements onto the plurality of paths in accordance with one or more end to end control adjustments;

wherein the end to end control adjustments include a server selection adjustment, an internet service provider (ISP) selector, and an overlay routing adjustment;

the server selection adjustment permitting selection of a server from a plurality of servers for each of a plurality of clients based upon a traffic demand for each of the plurality of clients and an available bandwidth between each of the plurality of clients and each of the plurality of servers, wherein the server selection adjustment determines one server from the plurality of servers for one client of the plurality of clients;

the ISP selector permitting selection of a respective ISP for each of a plurality of corresponding client/server requests; and

the overlay routing adjustment mapping a traffic demand for an application among a set of endpoints on the network.

3. The method of claim 1, further comprising determining traffic demand between the set of endpoints by utilizing a flow conservation constraint and edge capacity constraint;

wherein the flow conservation constraint comprises capturing routing choices;

wherein the flow conservation constraint is defined by a matrix I where I_ij=1 if edge j enters node i, and where I_ij=−

1 if edge j leaves node i; and

wherein a flow column F is defined such that F_i represents a flow on edge i.

4. The method of claim 3, wherein the edge capacity constraint is Σ

_p * F_p<

B, where F_p represents a matrix of edge flows for a commodity p, and where B is an edge capacity vector.

5. The method of claim 4, wherein the flow conservation constraint for each flow is I.F_p=K;

wherein K is a vector with value A_ij for source node i and value −

A_ij for sink node j; and

wherein I.F is an array where the i^th entry is 0 if node i is an intermediate node.

2. A computer program product for managing application performance on a network, the computer program product comprising a storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for facilitating a method comprising:

generating a network graph from a set of application endpoints on the network, the network graph being indicative of a plurality of paths between the application endpoints, the plurality of paths each including a plurality of available resources on the network;

annotating the network graph by associating one or more of the application endpoints with a corresponding latency annotation and a corresponding available bandwidth annotation, wherein the latency annotation includes one or more application performance constraints and the available bandwidth notation includes one or more capacity constraints; and

formulating an end to end plan from the annotated network graph wherein the end to end plan maps one or more application requirements onto the plurality of paths in accordance with one or more end to end control adjustments;

wherein the end to end control adjustments include a server selection adjustment, an internet service provider (ISP) selector, and an overlay routing adjustment;

the server selection adjustment permitting selection of a server from a plurality of servers for each of a plurality of clients based upon a traffic demand for each of the plurality of clients and an available bandwidth between each of the plurality of clients and each of the plurality of servers, wherein the server selection adjustment determines one server from the plurality of servers for one client of the plurality of clients;

the ISP selector permitting selection of a respective ISP for each of a plurality of corresponding client/server requests; and

the overlay routing adjustment mapping a traffic demand for an application among a set of endpoints on the network.