PLEX86  x86- Virtual Machine (VM) Program
 CVS  |  Mailing List  |  Download  |  Newsgroups

virtual memory 4471


Your Ad Here

Your Ad Here

Brian Inglis

...

It's (as usual) actually more complicated than that.

While I can't remember exactly how VMS managed inter-process memory conflicts, NT and derivatives certainly attempt to do so with some intelligence. The whole 'working set' approach buttumes that the OS can ascertain (preferably dynamically based on the process' paging history) some minimum occupancy below which a given process won't be able to function efficiently, and use that information to let some minimum set of processes take the hit when physical memory gets tight rather than cripple *all* processes by taking some smaller amount of memory from each. When the mechanism works as intended, the result is that as many processes as possible continue to run (and run reasonably well), with bulk (and at least to some degree on-disk-contiguous) evictions and restorations making more efficient use of the system resources than a situation in which processes are allowed to engage in uncontrolled fine-grained global page-thrashing.

virtual memory 4473
ref: I posted it numerous times before ... so there were URLs to where it had been repeatedly posted before basically i...

In situations where physical memory is less constrained, the per-process working sets expand beyond their minimum values (again, preferably according to individual process needs rather than in a more static manner - in which case the result does start to approximate a global page-replacement policy, just one implemented in two tiers rather than directly).

virtual memory 4474
Anne & Lynn Wheeler After our previous discussions, I looked high and low for any details on the grenoble system, and fifo or local working sets...

The underlying rationale is sound, and used to good effect in some environments to handle portions of database queries (where it's sometimes somewhat easier to ascertain what a given operation may need to function well - such as an operation involving repeated bulk access to a smallish relation which would very much like to remain memory-resident until the operation completes): rather than let the query dive right into the operation and execute inefficiently, when sufficient RAM is likely to become available soon the system allows other activity (that it thinks *can* execute efficiently) to continue until sufficient RAM becomes available to allow the new operation to do so.

- bill

virtual memory 4472
I remember similar conclusions (a) from when I was at Gould 1985-1989, where we put the latest and greatest working set scheduler into UNIX - and then backed off (b...



Your Ad Here

List | Previous | Next

virtual memory 4472

Alt Folklore Computers from Newsgroups

The #1 Usenet Provider on the Internet

virtual memory 4470