Hi all I was wondering if you guys have an opinion on the following data communication methods. The data are mostly a bunch of binary where program1 is asking program2 to do something with the data and send the results back to program1. The data is currently very small less than 1Kbyte, but I am not sure if it will grow later. I am also not sure how much time program2 needs to send a response.
method 1 - This simply uses plumber and it greatly simplifies both program1 and program2 program1 --> file IO -> plumber -> program2 -> file IO -> plumber -> program1 method 2 - In this scheme program2 is a server, mounted on the file system. The client does not need to know 9p program1 -> file IO -> server/progam2 -> fileIO -> program1 method 3 - program2 is still a server but, the client uses 9p to communicate with program2 program1 -> 9p -> server/program2 -> 9p -> program1 I hope that made some sense to somebody. regards fernan -- http://www.fernski.com
