Description:
In DBT2 benchmarks, 10 warehouses (database size ~1GB)
4.1 13616.51 NOTPM
5.0-nonsp 12447.16 NOTPM (ration 0.91)
i.e. 5.0 is slower by ~9%
How to repeat:
DBT2-benchmark (http://www.osdl.org/lab_activities/kernel_testing/osdl_database_test_suite/osdl_dbt-2/)
run line:
./run_mysql.sh -c 20 -w 10 -t 300 -e
DATABASE CONNECTIONS: 20
TERMINAL THREADS: 100
SCALE FACTOR(WARHOUSES): 10
TERMINALS PER WAREHOUSE: 10
DURATION OF TEST(in sec): 180
SLEEPY in (msec) 300
ZERO DELAYS MODE: 1
mysql params:
PARAM="--port=3306 \
--socket=/tmp/mysql.sock \
--user=root \
--log-error=/benchmarks/work/err4.log \
--datadir=$FSPATH \
--basedir=$BASEPATH \
--max_connections=1600 \
--max_connect_errors=10 \
--table_cache=2048 \
--max_allowed_packet=1048576 \
--binlog_cache_size=1048576 \
--max_heap_table_size=67108864 \
--sort_buffer_size=65536 \
--join_buffer_size=1048576 \
--thread_cache=16 \
--thread_concurrency=16 \
--thread_stack=64K \
--query_cache_size=0 \
--ft_min_word_len=4 \
--default_table_type=MYISAM \
--transaction_isolation=REPEATABLE-READ \
--tmp_table_size=64M \
--skip-locking \
--server-id=1 \
--innodb_status_file=0 \
--innodb_data_home_dir=$FSPATH \
--innodb_data_file_path=ibdata1:100M:autoextend \
--innodb_log_group_home_dir=$FSPATH \
--innodb_buffer_pool_size=1300M \
--innodb_additional_mem_pool_size=20M \
--innodb_log_file_size=650M \
--innodb_log_files_in_group=2 \
--innodb_log_buffer_size=8M \
--innodb_flush_log_at_trx_commit=1 \
--innodb_lock_wait_timeout=300 \
--innodb_locks_unsafe_for_binlog=0 \
--innodb_thread_concurrency=1000 \
--innodb_flush_method=O_DIRECT "
