Browse Prior Art Database

Workload Balancing Algorithm to bias request routing based on client network latency/connection speed

IP.com Disclosure Number: IPCOM000235484D
Publication Date: 2014-Mar-04
Document File: 2 page(s) / 61K

Publishing Venue

The IP.com Prior Art Database

Abstract

A workload balancing algorithm to bias request routing based on client network latency/connection speed is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 63% of the total text.

Page 01 of 2

Workload Balancing Algorithm to bias request routing based on client network latency/connection speed

Disclosed is a workload balancing algorithm to bias request routing based on client network latency/connection speed.

In an enterprise data processing environment, system resources are often limited. Resources should be biased towards higher priority requests. In some circumstances, it may be logical that requests made over a connection of low connection speed can be handled with a lower priority than requests with an acceptable or fast connection.

Applications (for example, Java

                    ® enterprise applications) may run in a clustered environment for high availability and scalability. Disclosed is an algorithm to bias/route requests from slow connection speed clients to a specific clone or Java Virtual Machine (JVM) cluster member. The algorithm is based on sampling the time taken to receive content versus content length. This sampling may be performed by a Workload Balancing (WLB) device or webserver plugin.

The Figure below demonstrates that when a clients network latency becomes large, priority handling of the client request on the server can do little to change the observed client response time.

Observed client response time

Duration of client request handled by fast server (high network latency)

Duration of client request handled by slow server (high network latency)

Duration of client request handled by fast server (low network ...