[David Jeske <jeske@genesis.isdn.uiuc.edu> writes:]
>My question is, does the current VSTa system allow high-level servers to take
>lots of CPU running the requests of low level clients? If it does, then it
>certainly does not "fairly" share the CPU. QNX has a system for adjusting
>the priority of a server to match the priority of the client whose request
>it's serving. Is something like this worth looking into?
This is known as "back end promotion". When we looked at providing it at
HP, our ultimate decision was that its benefits were far outweighed by its
costs.
No system server uses "lots" of CPU (for some definition of "lots"). A low
priority client would be hard-pressed to do a CPU attack via a high priority
server, because the service itself would run promptly, but the low priority
client would be throttled by his need to run again to launch the next
request.
Note that VSTa was NOT designed as a real-time system. As I tried to draw
out in the overview paper, there are mechanisms often found in real-time
systems which are very useful in making a microkernel operate as a general-
purpose system. This is not to say I oppose using VSTa in RT, but it can
help you understand why some elements of a RT system have been much more
developed than others.
Andy
Received on Wed Dec 14 10:32:26 1994
This archive was generated by hypermail 2.1.8 : Thu Sep 22 2005 - 15:12:11 PDT