Article View: news.software.b
Article #4783Re: Multi-FS news spool layout question
From: Ken Hornstein
Date: Thu, 17 Dec 1992 01:32
Date: Thu, 17 Dec 1992 01:32
16 lines
1417 bytes
1417 bytes
In article <1gofqvINNee@milk.Warren.MENTORG.COM> tal@Warren.MENTORG.COM (Tom Limoncelli) writes: >In article <1gli95INNid5@cat.cis.Brown.EDU> JIM@brownvm.brown.edu (James Mathiesen) writes: > >>A more elegent solution is to let dbz() have a write-through cache, >>this lets .pag be kept in memory, but updates are written to disk >>as they occur. I'm enclosing patches below to dbz.c, dbz.h and >>his.c that add and use dbzwritethough(). > >Excuse my ignorance, but would it be better if the history.pag file was >mmap()'ed? that way, a simple fsync() call would write out just the >changes rather than the entire file. Yes, it would. Unfortunately, under Ultrix (which I believe what the original poster was talking about), you can only mmap() a character device. *sigh* --Ken
Technical Details
Message-ID:
<1golb8INNi8e@wintermute.phys.psu.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!ukma!psuvax1!atlantis.psu.edu!wintermute.phys.psu.edu!leps5.phys.psu.edu!kenhReferences:
<1glafaINN393@access.usask.ca> <1gli95INNid5@cat.cis.Brown.EDU> <1gofqvINNee@milk.Warren.MENTORG.COM>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!usenet.ins.cwru.edu!ukma!psuvax1!atlantis.psu.edu!wintermute.phys.psu.edu!leps5.phys.psu.edu!kenh From: kenh@leps5.phys.psu.edu (Ken Hornstein) Newsgroups: news.software.b Subject: Re: Multi-FS news spool layout question Date: 17 Dec 1992 01:32:24 GMT Organization: Penn State, Laboratory for Elementary Particle Science Lines: 16 Message-ID: <1golb8INNi8e@wintermute.phys.psu.edu> References: <1glafaINN393@access.usask.ca> <1gli95INNid5@cat.cis.Brown.EDU> <1gofqvINNee@milk.Warren.MENTORG.COM> Nntp-Posting-Host: leps5.phys.psu.edu