On Friday, November 7, 2003, at 03:06 AM, Jan Pieter Kunst wrote:
Hardware: 1.8 GHz G5 with 1GB RAM
Panther (Mac OS X 10.3) doesn't have the sudden slowness in the 'insert' test that Jaguar (Mac OS X 10.2) has. Total time for Panther is faster than total time for Jaguar (5561.00 secs vs. 6770.00 secs). 'Insert' is faster on Panther than on Jaguar (3316 secs vs. 5215 secs), but all other benchmarks are somewhat slower on Panther. I'm a bit worried about the 'select', which is about twice as slow on Panther than on Jaguar (1428 secs vs. 743 secs).
Complete benchmark results below, sorry for the possible bandwidth waste, but I'm not sure what I could snip, if anything.
JP
Panther benchmarks:
======================================================================= ======
Benchmark DBD suite: 2.14
Date of test: 2003-11-07 0:12:11
Running tests on: Darwin 7.0.0 Power Macintosh
Arguments:
Comments:
Limits from: mysql,pg,solid
Server version: MySQL 4.0.15 standard log
Optimization: None
Hardware:
alter-table: Total time: 141 wallclock secs ( 0.22 usr 0.11 sys + 0.00 cusr 0.00 csys = 0.33 CPU)
ATIS: Total time: 59 wallclock secs (10.59 usr 6.30 sys + 0.00 cusr 0.00 csys = 16.89 CPU)
big-tables: Total time: 57 wallclock secs ( 7.53 usr 13.50 sys + 0.00 cusr 0.00 csys = 21.03 CPU)
connect: Total time: 252 wallclock secs (44.31 usr 38.81 sys + 0.00 cusr 0.00 csys = 83.12 CPU)
create: Total time: 298 wallclock secs ( 8.76 usr 3.06 sys + 0.00 cusr 0.00 csys = 11.82 CPU)
insert: Total time: 3316 wallclock secs (594.82 usr 261.66 sys + 0.00 cusr 0.00 csys = 856.48 CPU)
select: Total time: 1428 wallclock secs (67.50 usr 25.18 sys + 0.00 cusr 0.00 csys = 92.68 CPU)
transactions: Test skipped because the database doesn't support transactions
wisconsin: Total time: 27 wallclock secs ( 4.19 usr 2.58 sys + 0.00 cusr 0.00 csys = 6.77 CPU)
I'm not sure how useful it is for comparison, but our dual 2GHz G5 w/ 4GB of RAM had these results (for MyISAM tables):
alter-table: Total time: 11 wallclock secs ( 0.02 usr 0.04 sys + 0.00 cusr 0.00 csys = 0.06 CPU)
ATIS: Total time: 21 wallclock secs ( 9.41 usr 9.44 sys + 0.00 cusr 0.00 csys = 18.85 CPU)
big-tables: Total time: 34 wallclock secs ( 7.90 usr 20.86 sys + 0.00 cusr 0.00 csys = 28.76 CPU)
connect: Total time: 180 wallclock secs (45.56 usr 51.51 sys + 0.00 cusr 0.00 csys = 97.07 CPU)
create: Total time: 118 wallclock secs ( 3.87 usr 2.55 sys + 0.00 cusr 0.00 csys = 6.42 CPU)
insert: Total time: 1437 wallclock secs (432.84 usr 358.12 sys + 0.00 cusr 0.00 csys = 790.96 CPU)
select: Total time: 146 wallclock secs (42.52 usr 33.74 sys + 0.00 cusr 0.00 csys = 76.26 CPU)
transactions: Test skipped because the database doesn't support transactions
wisconsin: Total time: 12 wallclock secs ( 2.91 usr 2.76 sys + 0.00 cusr 0.00 csys = 5.67 CPU)
How the select portion of the test could be 10x slower, I'm not sure. Have you tuned MySQL at all? key_buffer, etc?
- Gabriel
-- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
