eugene@eos.arc.nasa.gov (Eugene Miya) (12/20/88)
I think the question of asynchonrous and intrusive effects will only be solved when computists get better chances to work on highly instrumented stand alone machines. Cache line effects, paging, etc. must have the potential to be systematically eliminated. It's almost a joke (non-scientific) to be doing some of this research on the machines we have today. Compilers have to be made more portable to compare effects, applications should be given a chance to run without the OS (or the OS checking to see if other processes exist). We must be give the chance to separate and compare. Do National computing research labs need to exist like telescopes and accelerators? Perhaps. Fortunately we have networks, this helps some. There's too many economic questions which would cloud this type of issue. Another gross generalization from --eugene miya, NASA Ames Research Center, eugene@aurora.arc.nasa.gov resident cynic at the Rock of Ages Home for Retired Hackers: Sorry, posting too much