Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!iuvax!purdue!haven!mimsy!chris From: chris@mimsy.UUCP (Chris Torek) Newsgroups: comp.arch Subject: Re: Memory utilization Message-ID: <19392@mimsy.UUCP> Date: 2 Sep 89 23:31:29 GMT References: <3332@blake.acs.washington.edu> <261500008@S34.Prime.COM> Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742 Lines: 19 In article pcg@thor.cs.aber.ac.uk (Piercarlo Grandi) writes: >The BSD pager was designed very very well by an extremely >competent person with a specialization in performance analysis of >virtual memory; unfortunately it was developed on a machine with >512Kbytes. On today's machines with 8-16-32 Mbytes, a clock scan >may take several minutes, making the well designed algorithm >essentially useless. A good example of how unanticipated changes >of scale make algorithms inappropriate. It was retuned fairly early for a 2 MB machine (probably ernie or ucbvax), and the 2 MB version is what was in 4.1 and 4.2 BSD. Since then it has been redone as a two-handed clock (one for simulating reference, one for freeing pages) and the hands set 2 MB apart. That helps a great deal, but the system still does not really deal well with heavy paging loads, mostly because DEC disks are too slow. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163) Domain: chris@mimsy.umd.edu Path: uunet!mimsy!chris