hidinger@cod.NOSC.MIL (Ronald M. Hidinger) (01/26/88)
We are interested in doing some realtime continuous system
simulation with Apollos for systems with bandwidths around 100 hz.
A quick look through my manuals left me with the following questions:
1. Is there any way to control process priority from within the process?
(Other than the ever useful pgm_$invoke("/com/ppri"...))
2. Is there any way to lock a process in memory?
3. What methods of time synchronization are available? The system timer
event count has a resolution of only 0.25 s; not nearly good enough.
Time_$wait has the resolution, but precludes answering other events.
(Interactive user input is required.) It appears the only option is
a user defined timing event count.
4. What sort of relative computation load can we get away with and still
get good realtime response?
5. How much network or local I/O activity can be tolerated?
6. Can the priority of OS processes be controlled?
Send me your experiences, opinions, etc. and I'll summarize.
Ron Hidinger
Naval Ocean Systems Center, San Diego, CA 92152
hidinger@nosc.mil
sdcsvax!noscvax!hidinger