Lowest latency programming on Linux, where to start?

I heard that there are some ways to change linux in such a way that a particular application can get a very low latency, so when it requests resources, the OS will try to provide it with a resource as soon as possible by canceling the default priority multitasking mechanism, I don't have CS background but the application I'm working on is very sensitive to delays, can someone tell me if there are any documents / materials on this particular issue? thank you very much.

+6
source share
2 answers

A guaranteed response with a low latency is called real time . This means that the guaranteed deadlines for achieving the goal are guaranteed.

There is a project called RTLinux. See the Real-Time Linux Wiki: https://rt.wiki.kernel.org/index.php/Main_Page

+4
source

There are two real-time models:

  • soft real-time system - you get it using RT preempt kernel patches . I think this ensures context transition within 10 ms. The goal of this project is to meet the stringent requirements of real time.
  • rigid real-time system - have more stringent guarantees (1 ms response). There are several libraries (e.g. xenomai ) that claim to provide a rigid real-time system.
+1
source

All Articles