tom@iconsys.UUCP (Tom Kimpton) (03/05/88)
In article <219@eos.UUCP> jaw@eos.UUCP (James A. Woods) writes: [stuff deleted related to disk I/O timings ] >for the mac2, these numbers harden the very rough timing for > time cat /usr/dict/words > /dev/null > >(~200K bytes in four CPU seconds) done on an beta a/ux a year ago. it >looks almost entirely due to the speed of the system5 vs. bsd4.2 filesystem. >considering how much other bsd stuff apple picked up, >why they didn't go for the berkeley filesystem is beyond me ... One reason could be the potentiality of breaking programs that deal directly with the file system. I'm probably talking off the top of my head because I don't know any about Apple's implementation, but in generic SYSV the inodes are all grouped together at the "head" of the disk, with the data blocks following, while on BSD4.2 inodes are at the "head" of each cylinder group (i.e. scattered in chunks in the midst of the data blocks). Thus programs (like "volcopy", or "fsdb") that work on a basic assumption about the disk structure need a major amount of work to make them work on the new file system. Just one less headache in the midst of a UN*X port migraine. Maybe when it settles down, they'll change the file system. -- Tom Kimpton {ihnp4,uunet}!iconsys!tom Icon International, Inc. {ihnp4,psivax}!nrcvax!nrc-ut!iconsys!tom Orem, Utah 84058 ARPANET: icon%byuadam.bitnet@wiscvm.wisc.edu (801) 225-6888 BITNET: icon%byuadam.bitnet