Hello benchmark fans,
100 warehouse runs did not run properly on caneland,
lu0009, and nehalem-1
News for this week:
* I tested a patch from Kevin on caneland, which gave us
40 - 60% performance gain for Falcon.
* caneland did not run 100 warehouses properly
due to OS issues.
caneland.mysql.com[1] numbers can be found here:
https://intranet.mysql.com/~hakank/caneland/weekly_dbt2_overview/weekly_dbt2_overview.php
lu0009.mysql.com[2] numbers can be found here:
https://intranet.mysql.com/~hakank/lu0009/weekly_dbt2_overview/weekly_dbt2_overview.php
nehalem-1.mysql.com[3] numbers can be found here:
https://intranet.mysql.com/~hakank/nehalem-1/weekly_dbt2_overview/weekly_dbt2_overview.php
walldorf@home[4] numbers can be found here:
https://intranet.mysql.com/~hakank/walldorf/weekly_dbt2_overview/weekly_dbt2_overview.php
Brief overview
****************
*** caneland ***
****************
-w10
-c8 -c16 -c32 -c64 -c99
F/I 37.36% 29.62% 31.60% 30.85% 29.62%
-w20
-c8 -c16 -c32 -c64 -c128 -c192
F/I 37.35% 32.11% 35.53% 34.05% 30.23% 32.35%
-w100
-c8 -c16 -c32 -c64 -c128 -c192 -c256 -c499
F/I 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
**************
*** lu0009 ***
**************
-w10
-c8 -c16 -c32 -c64 -c99
F/I 84.87% 53.57% 38.71% 37.00% 35.49%
-w20
-c8 -c16 -c32 -c64 -c128 -c192
F/I 65.80% 59.37% 53.12% 50.59% 52.55% 62.81%
-w100
-c8 -c16 -c32 -c64 -c128 -c192 -c256 -c499
F/I 76.07% 74.53% 0.00% 0.00% 54.93% 0.00% 53.29% 0.00%
*****************
*** nehalem-1 ***
*****************
-w10
-c8 -c16 -c32 -c64 -c99
F/I 63.00% 66.20% 60.52% 60.59% 67.04%
-w20
-c8 -c16 -c32 -c64 -c128 -c192
F/I 35.00% 46.67% 45.73% 46.03% 55.91% 66.38%
-w100
-c8 -c16 -c32 -c64 -c128 -c192 -c256 -c499
F/I 0.00% 0.00% 0.00% 0.00% 95.98% 0.00% 0.00% 99.92%
****************
*** walldorf ***
****************
-w10
-c8 -c16 -c32 -c64 -c99
F/I 61.35% 57.90% 55.94% 54.68% 53.46%
-w20
-c8 -c16 -c32 -c64 -c128 -c192
F/I 54.27% 50.11% 53.65% 51.93% 52.34% 53.15%
-w100
-c8 -c16 -c32 -c64 -c128 -c192 -c256 -c499
F/I 93.83% 113.25% 131.61% 133.05% 137.52% 143.66% 151.36% 144.15%
Todo:
- Put temporary results directory of DBT2 to local disk
instead of storing the data NFS.
- Remove results from crashed runs in mean value calculation.
- Add more information from INFORMATION_SCHEMA tables like Falcon's
system memory usage or summary of active/pending_commit/pending_completion.
Also talk to Ann to get more insight to Falcon metrics.
- On walldorf put falcon_serial_log_dir to SATA disk to see
whether the battery backed up RAID controller is the source of
huge performance gain we saw lately.
- Get a Solaris SPARC and Intel/AMD machine for DBT2.
Change set used:
revno: 2700
revision-id: hky@stripped
parent: hky@stripped
committer: Hakan Kuecuekyilmaz
branch nick: mysql-6.0-falcon-to-merge
timestamp: Thu 2009-04-30 12:33:34 +0200
Setup: datadir is on RAID10 and logdir is on separate disk (for InnoDB).
Configuration is:
MYSQLD_CONFIG_BASE="--no-defaults \
--socket=/tmp/mysql.sock"
MYSQLD_CONFIG_GENERAL="$MYSQLD_CONFIG_BASE \
--basedir=$FALCON \
--datadir=$DATA_DIR \
--tmpdir=/tmp \
--language=${FALCON}/sql/share/english \
--key_buffer_size=8M \
--skip-grant-tables \
--sort_buffer_size=2M \
--read_buffer_size=1M \
--table_open_cache=5120 \
--tmp_table_size=67108864 \
--thread_cache=1024 \
--query_cache_size=0 \
--query_cache_type=0 \
--max_allowed_packet=512M \
--max_connections=707 \
--log-error"
MYSQLD_CONFIG_FALCON="$MYSQLD_CONFIG_GENERAL \
--skip_innodb \
--default_storage_engine=Falcon \
--falcon_io_threads=10 \
--falcon_record_memory_max=650M \
--falcon_page_cache_size=1650M \
--falcon_page_size=8k \
--falcon_serial_log_dir=$DATA_DIR \
--falcon_serial_log_buffers=100 \
--falcon_consistent_read=off"
MYSQLD_CONFIG_INNODB="$MYSQLD_CONFIG_GENERAL \
--skip_falcon \
--default_storage_engine=InnoDB \
--innodb_data_home_dir=$DATA_DIR \
--innodb_data_file_path=ibdata1:10M:autoextend \
--innodb_log_group_home_dir=$LOG_DIR \
--innodb_buffer_pool_size=2300M \
--innodb_additional_mem_pool_size=20M \
--innodb_log_file_size=650M \
--innodb_log_buffer_size=16M \
--innodb_lock_wait_timeout=50 \
--innodb_file_io_threads=10 \
--innodb_flush_log_at_trx_commit=1 \
--innodb_flush_method=O_DIRECT \
--innodb_log_files_in_group=6 \
--innodb_support_xa=0 \
--innodb_doublewrite=0 \
--innodb-status-file=0 \
--skip-innodb-adaptive-hash-index \
--innodb_max_dirty_pages_pct=70 \
--innodb_thread_concurrency=0"
[1] caneland.mysql.com Fedora release 10
Quad HexaCore @ 2.9GHz. Intel Xeon with 16MB L2 cache, 32GB RAM, hardware
RAID 10 on 4 x 15k RPM SAS disks.
[2] lu0009.mysql.com Ubuntu 8.10
Quad DualCore @ 3.4Ghz. Intel Xeon (with hyperthreading) with 16MB L2 cache, 8GB RAM,
hardware RAID 10 on 4 x 15k RPM SAS disks. This machine has unofficial Intel CPUs.
[3] nehalem-1.mysql.com Ubuntu 8.10
Dual DualCore @ 2.4Ghz. Intel Xeon (with SMT) with 16MB L2 cache, 16GB RAM,
RAID10 on 4 x SATA disk.
[4] walldorf@home RHEL 5
Dual DualCore @ 2Ghz. Intel Xeon with 4MB L2 cache, 8GB RAM, hardware
RAID 10 on 4 x 15k RPM SAS disks.
Best regards,
Hakan
--
Hakan Küçükyılmaz, Senior Software Engineer DBTG/MySQL +49 160
98953296
Sun Microsystems GmbH Sonnenallee 1, DE-85551 Kirchheim-Heimstetten
Geschaeftsfuehrer: Thomas Schroeder, Wolfang Engels, Dr. Roland Boemer
Vorsitz d. Aufs.rat.: Martin Haering HRB MUC 161028 49.011, 8.376
| Thread |
|---|
| • Weekly Falcon/InnoDB DBT2 Overview 2009-05-04 | Hakan Kuecuekyilmaz | 4 May |
