I am working on Processor-Pool model for a project. Here are some assumptions:
1. Users have a low-end system.
2. They want to use an application which requires high-end workstations.
3. They need real-time graphic(visual) updates.
I was wondering if it is possible to do this without screen capture or TeamViewer, or RealVNC, or Red5 like software. I came across an idea to develop a network display model through the following method:
a. The user has a small software installed on their system which is connected via sockets to the server. Each user input is sent to the server(This is where Processor Pool model come into play).
b. Like cloud gaming services, the input is processed and the display is generated. But instead of sending video output we generate a file which is sent to the user.
c. The user receives this file and the small software installed on his system generates the final display output.
Now my question is what might be the extension of file highlighted above. Conventionally we have the file with extensions like JPG/ AAC/ H.265 codec/ etc.
Is it possible to generate a JSON for the file and which when received by the software at the user end generates the desired display output?
Please post your queries, if you have any. I am looking forward to discussing with all the great people out there.
Note: I am not looking for Screen Capture solutions. And a big TIA
You can use java RMI and Corba application to do this thing. Further in distributed system you can send bulk data to different high end distributed server and you can get output to any lightweight system connected in the distributed.
Have CORBA and Java RMI been used for GUI based applications where the sent data was used for generating GUI for high-end application?
I gave a general overview for finding out if they have been used for GUI based application but couldn't find anything.
TIA
I believe that you are trying to reinvent something that already exists, which is fine! But, in this particular instance, VNC as a server-client method is pretty much what you ought to find sufficient for your explained use-case scenario.
The server can run a virtual screen which renders the application that demands a a lot of processing power to run (which the serve supplies one way or another) and the client is just a dumb terminal capable of performing minimal data decoding at some phase and showing it on the physical screen.
The method you've proposed (using JSON and JPEG) suggests that you want to send frame-by-frame data, which can be troublesome in terms of bandwidth, required down-link speed and inability to perform multi-frame compression (between key-frames).
VNC allows all of this. It is efficient, optimised, has different modes for different targeted frame-rates and display content change dynamics.
Your proposed method would have to be designed and developed from scratch, which is, in this particular instance, probably a waste of your time and resources.
Of course, VNC is not the only possible solutions. There are others, but they are all probably much better suited than an own (new) solution, unless the application itself (that will be ran on the server) can directly create a stream of video data and push it to the client, completely skipping the need for a virtual display. This, however, requires a specialised application made specifically to function that way.
- Badges
- Science method