There are two methods for getting data into a Thin Client system – through the Server, or through the Clients.
Getting data through the Server
For small installations, where input/output devices are not very far from the Server, and there are not many devices to read, this can be a very good solution. It is certainly the most traditional, with I/O devices being connected to serial ports or special hardware installed in the Server. As all Thin Client software runs on the Server anyway, there is no problem with each client identifying its input or output devices with physical addresses on the Server. This is also very efficient, as data doesn’t have to travel through the Client, to the Server, and back to the Client. However, it is very limited – a Server can only handle so much I/O, and if devices are far removed there will be difficulties in moving data (especially serial data) over the distance.