The architecture for a Thinfinity® Remote Desktop Server deployment is composed of:
- a Server Machine running Thinfinity® Remote Desktop Services
- a Server Machine running Gateway Services
- Thinfinity® Remote Desktop Web Client
(which loads on an HTML5 browser)
Thinfinity® Remote Desktop Server is a secure, high-performance HTTP / WebSockets server, which serves the web pages needed to run the Thinfinity® Remote Desktop Web Client on the web browser.
Thinfinity® Gateway Services respond to all web-page requests and, when a connection is solicited, it selects the appropriate Server to forward that request to. In case any established connection fails, or a Server falls down, the Gateway will be able to reconnect to the Server that has the highest availability at the moment.
When the end-user accesses the Thinfinity® Remote Desktop main page and enters the appropriate connection parameters, the Thinfinity® Remote Desktop Web Client communicates with the Gateway Services, which, in turn, forwards this connection to the Thinfinity®Remote Desktop Server, using Ajax and WebSockets (if available) to initiate a connection to the remote-end.
Once the connection is established, the server will receive RDP commands, optimize them for the web, and send the resulting data stream to the Gateway, which will send it to the Thinfinity® Remote Desktop Web Client.
If you arrive to the conclusion that your Thinfinity® Remote Desktop Server environment would benefit from using load balancing, you can choose between two possible architectures. This decision is an essential step in planning the hardware scheme and configuring the system to work in a distributed way.
Scenario 1: One Gateway and multiple Servers
In this simple scenario, a single Gateway distributes the connection load between a number of Servers.
Scenario 2: Multiple Gateways and multiple Servers
This second scheme is composed by multiple Servers, multiple Gateways and the DNS Server, its domain name associated to all the available Gateways’ IPs.