2025-02-04, 05:20 PM
lets take the following example
1 server for movies using built in igpu
1 server for tv shows using built in igpu
1 server for anime using built in igpu. Reality is it doesn't matter if its an igpu or sometime like a nvidia card
clients connect to the head end server, this server then communicated with the sub servers. If I want to watch a movie, the movie server does the transcoding and sends the data to the head server or the client. Tv show the tv server does the work and so on.
Instead of relying on one server to host the files and transcode you split the load between multiple machines but the client thinks its just a single server.
Alternatively using multiple transcoding hardware as well. Igpu/nvidia gpu at the same time.
1 server for movies using built in igpu
1 server for tv shows using built in igpu
1 server for anime using built in igpu. Reality is it doesn't matter if its an igpu or sometime like a nvidia card
clients connect to the head end server, this server then communicated with the sub servers. If I want to watch a movie, the movie server does the transcoding and sends the data to the head server or the client. Tv show the tv server does the work and so on.
Instead of relying on one server to host the files and transcode you split the load between multiple machines but the client thinks its just a single server.
Alternatively using multiple transcoding hardware as well. Igpu/nvidia gpu at the same time.