Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.csd.uwm.edu!cs.utexas.edu!usc!henry.jpl.nasa.gov!elroy.jpl.nasa.gov!ames!amdahl!mat From: mat@uts.amdahl.com (Mike Taylor) Newsgroups: comp.arch Subject: Re: Memory utilization & inter-process contention Message-ID: Date: 23 Aug 89 15:01:16 GMT References: <3332@blake.acs.washington.edu> <1989Aug22.163100.25540@utzoo.uucp> <9aid02rf4dNn01@amdahl.uts.amdahl.com> Organization: Amdahl Corporation, Sunnyvale CA Lines: 22 In article <9aid02rf4dNn01@amdahl.uts.amdahl.com>, sbf10@uts.amdahl.com (Samuel Fuller) writes: > > This type of job scheduling is what IBM's MVS operating system does > very well and what Unix does very poorly. As far as I know the > Unix operating system has no concept of job classes. > > Unix has the opposite problem. When a Unix system is very heavily loaded > everybody notices. Programs run slowly and response time goes to hell. > > Is there a happy middle ground? There is NQS, which adds some of the batch queuing ideas to Unix systems. What doesn't exist, as far as I know, is much control of dynamic resource allocation. MVS went through an evolution from manual control to automated control (SRM - System Resource Manager) which Unix has yet to follow. But even SRM cannot provide a free lunch - it attempts to avoid overcommitting resources by shutting off or reducing service to some users according to rules that you provide. -- Mike Taylor ...!{hplabs,amdcad,sun}!amdahl!mat [ This may not reflect my opinion, let alone anyone else's. ]