12 个解决方案
#1
看看最耗资源的SQL的命中率怎么样?
#2
怎么查?
#3
命中率是什么意思?命中率影响什么?
#4
SELECT EXECUTIONS , DISK_READS, BUFFER_GETS,这是看命中率小于80%的SQL
round((BUFFER_GETS-DISK_READS)/BUFFER_GETS,2) Hit_radio,
ROUND(DISK_READS/EXECUTIONS,2) Reads_per_run,
SQL_TEXT
FROM V$SQLAREA
WHERE EXECUTIONS>0
AND BUFFER_GETS > 0
AND (BUFFER_GETS-DISK_READS)/BUFFER_GETS < 0.8
ORDER BY 2 desc,4 DESC;
#5
你的数据库版本? 已经内存结构的配置?
#6
数据库高速缓存命中百分比受哪些因素影响?
#7
一个是shared_pool_size和db_cache_size的大小会影响命中率,一般是越大命中率越高,但是太大会增加寻址困难,10g的话还要看下sga_target这个参数.
另一个是sql的写法,尤其是没有使用绑定变量的sql会大大降低命中率
#8
附上spfile参数信息:
aq_tm_processes = 1
background_dump_dest = /orapp/admin/fun/bdump
cluster_database = TRUE
cluster_database_instances = 2
cluster_interconnects = 172.16.10.1
compatible = 9.2.0.0.0
control_files = ('/dev/rscz_ctl1_110m', '/dev/rscz_ctl2_110m', '/dev/rscz_ctl3_110m')
core_dump_dest = /orapp/admin/fun/cdump
cpu_count = 4
db_block_size = 8192
db_cache_size = 1073741824
db_domain = ''
db_file_multiblock_read_count = 16
db_name = fun
dispatchers = '(PROTOCOL=TCP) (SERVICE=funXDB)'
fast_start_mttr_target = 300
hash_join_enabled = TRUE
instance_name = fun1
instance_number = 1
java_pool_size = 318767104
job_queue_processes = 10
large_pool_size = 318767104
local_listener = 'LISTENER_FUN1'
log_archive_dest_1 = 'LOCATION=/ARCH'
log_archive_format = '%t_%s.dbf'
log_archive_start = TRUE
log_buffer = 5242880
nls_comp =
open_cursors = 10000
pga_aggregate_target = 524288000
processes = 512
query_rewrite_enabled = FALSE
remote_listener = LISTENERS_FUN
remote_login_passwordfile = EXCLUSIVE
service_names = fun
session_cached_cursors = 200
sessions = 1000
sga_max_size = 4301232496
shared_pool_size = 637534208
sort_area_size = 20242880
spfile = /dev/rscz_spf_5m
star_transformation_enabled = FALSE
thread = 1
timed_statistics = TRUE
undo_management = AUTO
undo_retention = 10800
undo_tablespace = UNDOTBS1
user_dump_dest = /orapp/admin/fun/udump
#9
简单介绍以下你的数据库配置:
SGA:
PGA:
数据库类型:OLTP/OLAP/DSS
数据库规模:xxxG
另外,命中率低和高都不能说明你的性能好和不好,核心是缓存最需要缓存的数据.
SGA:
PGA:
数据库类型:OLTP/OLAP/DSS
数据库规模:xxxG
另外,命中率低和高都不能说明你的性能好和不好,核心是缓存最需要缓存的数据.
#10
pfile 里加sga_target参数,让他自动管理,shared_pool_size,database buffer这类的就动态的进行管理了
*.sga_target=9868514300
只要sga_target < sga_max_size 就可以了
*.sga_target=9868514300
只要sga_target < sga_max_size 就可以了
#11
应该是9i的库,还没有SGA_TARGET
#12
简单介绍以下你的数据库配置: SGA: PGA:见引用;
数据库规模:40G
数据库版本:9.2.0.7
#1
看看最耗资源的SQL的命中率怎么样?
#2
怎么查?
#3
命中率是什么意思?命中率影响什么?
#4
SELECT EXECUTIONS , DISK_READS, BUFFER_GETS,这是看命中率小于80%的SQL
round((BUFFER_GETS-DISK_READS)/BUFFER_GETS,2) Hit_radio,
ROUND(DISK_READS/EXECUTIONS,2) Reads_per_run,
SQL_TEXT
FROM V$SQLAREA
WHERE EXECUTIONS>0
AND BUFFER_GETS > 0
AND (BUFFER_GETS-DISK_READS)/BUFFER_GETS < 0.8
ORDER BY 2 desc,4 DESC;
#5
你的数据库版本? 已经内存结构的配置?
#6
数据库高速缓存命中百分比受哪些因素影响?
#7
一个是shared_pool_size和db_cache_size的大小会影响命中率,一般是越大命中率越高,但是太大会增加寻址困难,10g的话还要看下sga_target这个参数.
另一个是sql的写法,尤其是没有使用绑定变量的sql会大大降低命中率
#8
附上spfile参数信息:
aq_tm_processes = 1
background_dump_dest = /orapp/admin/fun/bdump
cluster_database = TRUE
cluster_database_instances = 2
cluster_interconnects = 172.16.10.1
compatible = 9.2.0.0.0
control_files = ('/dev/rscz_ctl1_110m', '/dev/rscz_ctl2_110m', '/dev/rscz_ctl3_110m')
core_dump_dest = /orapp/admin/fun/cdump
cpu_count = 4
db_block_size = 8192
db_cache_size = 1073741824
db_domain = ''
db_file_multiblock_read_count = 16
db_name = fun
dispatchers = '(PROTOCOL=TCP) (SERVICE=funXDB)'
fast_start_mttr_target = 300
hash_join_enabled = TRUE
instance_name = fun1
instance_number = 1
java_pool_size = 318767104
job_queue_processes = 10
large_pool_size = 318767104
local_listener = 'LISTENER_FUN1'
log_archive_dest_1 = 'LOCATION=/ARCH'
log_archive_format = '%t_%s.dbf'
log_archive_start = TRUE
log_buffer = 5242880
nls_comp =
open_cursors = 10000
pga_aggregate_target = 524288000
processes = 512
query_rewrite_enabled = FALSE
remote_listener = LISTENERS_FUN
remote_login_passwordfile = EXCLUSIVE
service_names = fun
session_cached_cursors = 200
sessions = 1000
sga_max_size = 4301232496
shared_pool_size = 637534208
sort_area_size = 20242880
spfile = /dev/rscz_spf_5m
star_transformation_enabled = FALSE
thread = 1
timed_statistics = TRUE
undo_management = AUTO
undo_retention = 10800
undo_tablespace = UNDOTBS1
user_dump_dest = /orapp/admin/fun/udump
#9
简单介绍以下你的数据库配置:
SGA:
PGA:
数据库类型:OLTP/OLAP/DSS
数据库规模:xxxG
另外,命中率低和高都不能说明你的性能好和不好,核心是缓存最需要缓存的数据.
SGA:
PGA:
数据库类型:OLTP/OLAP/DSS
数据库规模:xxxG
另外,命中率低和高都不能说明你的性能好和不好,核心是缓存最需要缓存的数据.
#10
pfile 里加sga_target参数,让他自动管理,shared_pool_size,database buffer这类的就动态的进行管理了
*.sga_target=9868514300
只要sga_target < sga_max_size 就可以了
*.sga_target=9868514300
只要sga_target < sga_max_size 就可以了
#11
应该是9i的库,还没有SGA_TARGET
#12
简单介绍以下你的数据库配置: SGA: PGA:见引用;
数据库规模:40G
数据库版本:9.2.0.7