Browse Prior Art Database

Method for Load Balancing in Control Center Software Solutions

IP.com Disclosure Number: IPCOM000201597D
Original Publication Date: 2010-Nov-16
Included in the Prior Art Database: 2010-Nov-16
Document File: 4 page(s) / 829K

Publishing Venue

Siemens

Abstract

One of the major tasks of software used for controlling and managing control centers is the processing of real-time values received from field devices. Usually, real-time data processing is performed on real-time servers which receive the values coming from the field and run the applications for processing these incoming real-time values. Real-time servers are usually installed redundantly. One real-time server acts as so-called master and performs the processing of the incoming values. A second node, the so-called standby node, also retrieves the real-time values from the field. However, this node does not perform any data processing. Instead, it is used for redundancy, e. g., buffering all incoming values for a certain period of time, to be prepared in case of an outage of the master real-time server and to take over the tasks of the previous master real-time server. This configuration is depicted in Figure 1. Due to the actual processing of all incoming values on a single node, this configuration leads to disadvantages. High flow rates of incoming values, e. g., in case of emergency or outage, can create a very high amount of values to be processed on a single machine. This leads to a high load on the master real-time server, e. g., high CPU load (Central Processing Unit) or high memory consumption by single applications. Furthermore, a high load caused by applications processing real-time values causes other applications to lack in their performance, as the CPU load is consumed by other applications. This data lead to data loss or delayed data processing which is not acceptable if urgent or important messages, e. g., alarm messages, have to be delivered and displayed in the control centers.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 01 of 4

(This page contains 01 pictures or other non-text object)

(This page contains 00 pictures or other non-text object)

Method for Load Balancing in Control Center Software Solutions

Idea: Bernd Steiner, DE-Nuremberg

One of the major tasks of software used for controlling and managing control centers is the processing of real-time values received from field devices. Usually, real-time data processing is performed on real-time servers which receive the values coming from the field and run the applications for processing these incoming real-time values.

Real-time servers are usually installed redundantly. One real-time server acts as so-called master and performs the processing of the incoming values. A second node, the so-called standby node, also retrieves the real-time values from the field. However, this node does not perform any data processing. Instead, it is used for redundancy, e. g., buffering all incoming values for a certain period of time, to be prepared in case of an outage of the master real-time server and to take over the tasks of the previous master real-time server. This configuration is depicted in Figure 1. Due to the actual processing of all incoming values on a single node, this configuration leads to disadvantages. High flow rates of incoming values, e. g., in case of emergency or outage, can create a very high amount of values to be processed on a single machine. This leads to a high load on the master real-time server, e. g., high CPU load (Central Processing Unit) or high memory consumption by single applications. Furthermore, a high load caused by applications processing real-time values causes other applications to lack in their performance, as the CPU load is consumed by other applications. This data lead to data loss or delayed data processing which is not acceptable if urgent or important messages, e. g., alarm messages, have to be delivered and displayed in the control centers.

At present, applications which process real-time values coming from the field are improved by enabling them to process more real-time values with less CPU load or memory usage. This is reached by code optimization, improved algorithms or changes in the architecture of the control center software. However, the possibilities of these kinds of improvement are limited they can not be done endlessly. Thus, new solutions are needed to avoid the mentioned disadvantages.

In the following a novel method for load balancing in control center software solutions is proposed. Its configuration is depicted in Figure 2 and has the following key functionality:

• A new software component called Load Balancing Manager (LBM) is introduces. This component is labeled yellow in Figure 2.

• The LBM is intended to be running on a machine which may run at high load, e. g., a master real-time server, and on an additional balancing machine, e. g., a load balancing real-time server, with the same control center software installed on the master real-time

•...