I have a legacy application / database server that has very limited connectivity options.
It does have CGI built in, but CGI obviously has overheads. I can't use Fast-CGI as there are reliability issues.
Alternatively, a TCP port can be opened on the server, which can then handle HTTP requests.
However, the TCP port is not concurrent. Only one request can access the port at a time.
I have seen a proprietary solution that starts up a pool of TCP ports, and uses a custom Apache module to balance requests across the pool of TCP connections.
From an architectural point of view, I'm guessing this is no different to a pool of database connections, the only difference being that each connection is on a unique port.
As the Apache solution is proprietary, I am looking for an alternative solution.
Looking at the NGinx documentation, I think it is possible to set up a reverse proxy across many different ports, but would this work with single concurrency?
Would this be possible with configuration alone on NGinx?
Open to any other suggestions...
It does have CGI built in, but CGI obviously has overheads. I can't use Fast-CGI as there are reliability issues.
Alternatively, a TCP port can be opened on the server, which can then handle HTTP requests.
However, the TCP port is not concurrent. Only one request can access the port at a time.
I have seen a proprietary solution that starts up a pool of TCP ports, and uses a custom Apache module to balance requests across the pool of TCP connections.
From an architectural point of view, I'm guessing this is no different to a pool of database connections, the only difference being that each connection is on a unique port.
As the Apache solution is proprietary, I am looking for an alternative solution.
Looking at the NGinx documentation, I think it is possible to set up a reverse proxy across many different ports, but would this work with single concurrency?
Would this be possible with configuration alone on NGinx?
Open to any other suggestions...