Technical information
3/14
Modicon M580 automation 
platform
Ethernet Modbus/TCP network
Performance
Selecting the communication architecture
When selecting an architecture, performance needs to be taken into account at the 
earliest possible stage. To do this, developers need to:
1 Determine the precise requirements:
v quantity and type of devices to be interconnected
v volume and type of exchanges
v expected response times
v environment
2 Compare the requirements with the characteristics of the offers available and be 
aware that the actual performance level between any 2 points in an architecture 
depends on the weakest link in the chain, which can be:
v dependent on the hardware
v but also dependent on the applications (size, architecture, operating system, 
machine power rating, etc.), which are often only vaguely dened at this stage of the 
project
3 Decide which is the most suitable architecture
The purpose of the next few pages is to provide the main information and 
instructions needed to answer the second point. Given that the performance of an 
Ethernet architecture is linked to several parameters, these pages do not supply all 
the information needed to calculate the network performance. Their aim is to focus 
on the following main aspects:
b Guidelines for calculating the network load so as to design an Ethernet 
network that meets the application requirements
b Application response time to be obtained depending on the conguration used 
(see pages 3/15 to 3/17)
b Processing capability of Modicon M340, Modicon M580, Modicon Premium,
and Modicon Quantum platforms so as to be able to select the CPU and dene the 
number of Ethernet connections required on the PLC depending on the application 
(see pages 3/18 and 3/19)
Calculating the network load
Introduction
When calculating the load on an Ethernet network, all the communication services of 
all the peripheral devices connected to the network need to be calculated. 
Because of the outstanding performance of the Ethernet network, the load is often 
less than the Ethernet network limits and does not signicantly affect the application 
response time. This phenomenon is explained by the high speed of the Ethernet 
network: the network transaction time is 10% less than the application response 
time. In order to keep the network load low and avoid large theoretical calculations, 
it is highly advisable to separate the collision domain so as to limit the network load, 
using only the switched network (tree, star, or daisy-chain topology). 
Performance
Processors:
page 1/18
I/O architectures:
page 2/2
Software:
page 4/2
Ruggedized Modicon M580 
modules: page 5/2
2
1
3
4
5
6
7
8
9
10
2
1
3
4
5
6
7
8
9
10










