home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!Germany.EU.net!tools!fl
- From: fl@tools.de (Frank Lancaster)
- Newsgroups: comp.sys.acorn.tech
- Subject: Re: adfsbuffers
- Date: 23 Dec 92 15:14:47 GMT
- Organization: Tools GmbH, Bonn, Germany
- Lines: 21
- Message-ID: <FL.92Dec23141448@pierre.tools.de>
- References: <1992Dec18.113330@informatik.uni-kl.de><1992Dec20.011534.5366@cs.aukuni.ac.nz>
- <FL.92Dec21130356@pierre.tools.de>
- <1992Dec22.212234.2185@cs.aukuni.ac.nz>
- NNTP-Posting-Host: pierre.tools.de
- In-reply-to: jwil1@cs.aukuni.ac.nz's message of Tue, 22 Dec 1992 21:22:34 GMT
-
-
- In article <1992Dec22.212234.2185@cs.aukuni.ac.nz> jwil1@cs.aukuni.ac.nz (TMOTA) writes:
-
- > ... you don't think, perhaps, that the significant improvement is due to
- > the fact that you are reading multiple files and writing to 1? And
- > that the more buffers you have, the more files can be buffered before
- > buffers have to be flushed?
- > Remember that read-buffers should make a difference if you're using a
- > FS that supoprts read-ahead...
-
- No. I'm reading one file and writing another. The writing process is improved
- a lot. Especially performance when large files are written to the archive
- is improved. If I have set adfsbuffers to 255 and a file of 100KB is written to
- the archive, the writing routine returns immediately and the next file gets
- read and compressed while the writing process to the archive continues
- in the background.
-
- The C file routines obviously at some point get translated to OS_GPBP
- calls, so they are also influenced by adfsbuffers.
-
- -- Frank Lancaster, fl@tools.de
-