wiki:AppMultiThread

Multicore apps

Depending on your application and project, it may be desirable to develop a multi-thread application. Possible reasons to do this:

  • Your application's memory footprint is large enough that, on some PCs, there's not enough RAM to run a separate copy of the app on each CPU.
  • You want to reduce the turnaround time of your jobs (either because of human factors, or to reduce server occupancy).

You may be able to use OpenCL, MPI, OpenMP, CUDA, languages like Titanium or Cilk, or libraries of multi-threaded numerical "kernels", to develop a multi-threaded app.

Initialization

You will need to use the appropriate initialization function.

Waiting for children

If your application uses multiple process, the parent process must always wait for children to exit before exiting itself.

Thread priorities

You should set the priority of new threads to that of the main thread.

#ifdef MSVC
#define getThreadPriority() GetThreadPriority(GetCurrentThread())
#define setThreadPriority(num) SetThreadPriority(GetCurrentThread(), num)
#else
#include <sys/resource.h>
#define getThreadPriority() getpriority(PRIO_PROCESS, 0)
#define setThreadPriority(num) nice(num)
#endif

getThreadPriority() is called before forking, and setThreadPriority() is then called by each worker thread.

Specifying a plan class

Multicore apps must use the plan class mechanism to specify their properties. You may be able to use the predefined "mt" class. Otherwise you must specify your own, using either XML or C++.

Last modified 5 years ago Last modified on Jun 15, 2019, 6:53:56 PM