DB Performace & tuning

Question

Hello everyone,
I am just a beginner in Oracle DBA and I would like to know what are the best practices in tuning oracle databases?
I've read some documentation regarding DB performance and, for instance, I've found section about tuning PCT parameters for db blocks. I wonder if this could be done on production database without rebuilding tablespace??
Could give me some key points and advices (from real life) in oracle database tuning ?
Message was edited by:
Pawel S.

Answer 1

Hi,
You can also use this,
spool tuning_stats.txt
ttitle 'SYSTEM STATISTICS'
select 'LIBRARY CACHE STATISTICS:' from dual;
ttitle off
select 'PINS - # of times an item in the library cache was executed - '||
sum(pins),
'RELOADS - # of library cache misses on execution steps - '||
sum (reloads),
'RELOADS / PINS * 100 = '||round((sum(reloads) / sum(pins) *
100),2)||'%'
from v$librarycache
prompt Increase memory until RELOADS is near 0 but watch out for
prompt Paging/swapping
prompt To increase library cache, increase SHARED_POOL_SIZE
prompt
prompt ** NOTE: Increasing SHARED_POOL_SIZE will increase the SGA size.
prompt
prompt Library Cache Misses indicate that the Shared Pool is not big
prompt enough to hold the shared SQL area for all concurrently open cursors.
prompt If you have no Library Cache misses (PINS = 0), you may get a small
prompt increase in performance by setting CURSOR_SPACE_FOR_TIME = TRUE which
prompt prevents ORACLE from deallocating a shared SQL area while an
prompt application
prompt cursor associated with it is open.
prompt
prompt For Multi-threaded server, add 1K to SHARED_POOL_SIZE per user.
prompt
prompt------------------------------------------------------------------------
column xn1 format a50
column xn2 format a50
column xn3 format a50
column xv1 new_value xxv1 noprint
column xv2 new_value xxv2 noprint
column xv3 new_value xxv3 noprint
column d1 format a50
column d2 format a50
prompt HIT RATIO:
prompt
prompt Values Hit Ratio is calculated against:
prompt
select lpad(name,20,' ')||' = '||value xn1, value xv1
from v$sysstat
where name = 'db block gets'
select lpad(name,20,' ')||' = '||value xn2, value xv2
from v$sysstat
where name = 'consistent gets'
select lpad(name,20,' ')||' = '||value xn3, value xv3
from v$sysstat b
where name = 'physical reads'
set pages 60
select 'Logical reads = db block gets + consistent gets ',
lpad ('Logical Reads = ',24,' ')||to_char(&xxv1+&xxv2) d1
from dual
select 'Hit Ratio = (logical reads - physical reads) / logical reads',
lpad('Hit Ratio = ',24,' ')||
round( (((&xxv2+&xxv1) - &xxv3) / (&xxv2+&xxv1))*100,2 )||'%' d2
from dual
prompt If the hit ratio is less than 60%-70%, increase the initialization
prompt parameter DB_BLOCK_BUFFERS. ** NOTE: Increasing this parameter will
prompt increase the SGA size.
prompt
prompt------------------------------------------------------------------------
col name format a30
col gets format 9,999,999
col waits format 9,999,999
prompt ROLLBACK CONTENTION STATISTICS:
prompt
prompt GETS - # of gets on the rollback segment header
prompt WAITS - # of waits for the rollback segment header
set head on;
select name, waits, gets
from v$rollstat, v$rollname
where v$rollstat.usn = v$rollname.usn
set head off
select 'The average of waits/gets is '||
round((sum(waits) / sum(gets)) * 100,2)||'%'
From v$rollstat
prompt
prompt If the ratio of waits to gets is more than 1% or 2%, consider
prompt creating more rollback segments
prompt
prompt Another way to gauge rollback contention is:
prompt
column xn1 format 9999999
column xv1 new_value xxv1 noprint
set head on
select class, count
from v$waitstat
where class in ('system undo header', 'system undo block',
'undo header', 'undo block' )
set head off
select 'Total requests = '||sum(count) xn1, sum(count) xv1
from v$waitstat
select 'Contention for system undo header = '||
(round(count/(&xxv1+0.00000000001),4)) * 100||'%'
from v$waitstat
where class = 'system undo header'
select 'Contention for system undo block = '||
(round(count/(&xxv1+0.00000000001),4)) * 100||'%'
from v$waitstat
where class = 'system undo block'
select 'Contention for undo header = '||
(round(count/(&xxv1+0.00000000001),4)) * 100||'%'
from v$waitstat
where class = 'undo header'
select 'Contention for undo block = '||
(round(count/(&xxv1+0.00000000001),4)) * 100||'%'
from v$waitstat
where class = 'undo block'
prompt
prompt If the percentage for an area is more than 1% or 2%, consider
prompt creating more rollback segments. Note: This value is usually very
prompt small
prompt and has been rounded to 4 places.
prompt
prompt------------------------------------------------------------------------
prompt REDO CONTENTION STATISTICS:
prompt
prompt The following shows how often user processes had to wait for space in
prompt the redo log buffer:
select name||' = '||value
from v$sysstat
where name = 'redo log space requests'
prompt
prompt This value should be near 0. If this value increments consistently,
prompt processes have had to wait for space in the redo buffer. If this
prompt condition exists over time, increase the size of LOG_BUFFER in the
prompt init.ora file in increments of 5% until the value nears 0.
prompt ** NOTE: increasing the LOG_BUFFER value will increase total SGA size.
prompt
prompt -----------------------------------------------------------------------
col name format a15
col gets format 9999999
col misses format 9999999
col immediate_gets heading 'IMMED GETS' format 9999999
col immediate_misses heading 'IMMED MISS' format 9999999
col sleeps format 999999
prompt LATCH CONTENTION:
prompt
prompt GETS - # of successful willing-to-wait requests for a latch
prompt MISSES - # of times an initial willing-to-wait request was unsuccessful
prompt IMMEDIATE_GETS - # of successful immediate requests for each latch
prompt IMMEDIATE_MISSES = # of unsuccessful immediate requests for each latch
prompt SLEEPS - # of times a process waited and requests a latch after an
prompt initial willing-to-wait request
prompt
prompt If the latch requested with a willing-to-wait request is not
prompt available, the requesting process waits a short time and requests
prompt again.
prompt If the latch requested with an immediate request is not available,
prompt the requesting process does not wait, but continues processing
prompt
set head on
select name, gets, misses,
immediate_gets, immediate_misses, sleeps
from v$latch
where name in ('redo allocation', 'redo copy')
set head off
select 'Ratio of MISSES to GETS: '||
round((sum(misses)/(sum(gets)+0.00000000001) * 100),2)||'%'
from v$latch
where name in ('redo allocation', 'redo copy')
select 'Ratio of IMMEDIATE_MISSES to IMMEDIATE_GETS: '||
round((sum(immediate_misses)/
(sum(immediate_misses+immediate_gets)+0.00000000001) * 100),2)||'%'
from v$latch
where name in ('redo allocation', 'redo copy')
prompt
prompt If either ratio exceeds 1%, performance will be affected.
prompt
prompt Decreasing the size of LOG_SMALL_ENTRY_MAX_SIZE reduces the number of
prompt processes copying information on the redo allocation latch.
prompt
prompt Increasing the size of LOG_SIMULTANEOUS_COPIES will reduce contention
prompt for redo copy latches.
rem
rem This shows the library cache reloads
rem
set head on
prompt
prompt------------------------------------------------------------------------
prompt
prompt Look at gethitratio and pinhit ratio
prompt
prompt GETHITRATIO is number of GETHTS/GETS
prompt PINHIT RATIO is number of PINHITS/PINS - number close to 1 indicates
prompt that most objects requested for pinning have been cached. Pay close
prompt attention to PINHIT RATIO.
prompt
column namespace format a20 heading 'NAME'
column gets format 99999999 heading 'GETS'
column gethits format 99999999 heading 'GETHITS'
column gethitratio format 999.99 heading 'GET HIT|RATIO'
column pins format 9999999 heading 'PINHITS'
column pinhitratio format 999.99 heading 'PIN HIT|RATIO'
select namespace, gets, gethits,
gethitratio, pins, pinhitratio
from v$librarycache
rem
rem
rem This looks at the dictionary cache miss rate
rem
prompt
prompt------------------------------------------------------------------------
prompt THE DATA DICTIONARY CACHE:
prompt
prompt
prompt Consider keeping this below 5% to keep the data dictionary cache in
prompt the SGA. Up the SHARED_POOL_SIZE to improve this statistic. **NOTE:
prompt increasing the SHARED_POOL_SIZE will increase the SGA.
prompt
column dictcache format 999.99 heading 'Dictionary Cache | Ratio %'
select sum(getmisses) / (sum(gets)+0.00000000001) * 100 dictcache
from v$rowcache
prompt
prompt------------------------------------------------------------------------
prompt
prompt SYSTEM EVENTS:
prompt
prompt Not sure of the value of this section yet but it looks interesting.
prompt
col event format a37 heading 'Event'
col total_waits format 99999999 heading 'Total|Waits'
col time_waited format 9999999999 heading 'Time Wait|In Hndrds'
col total_timeouts format 999999 heading 'Timeout'
col average_wait heading 'Average|Time' format 999999.999
set pages 999
select *
from v$system_event
prompt
prompt------------------------------------------------------------------------
rem
rem
rem This looks at the sga area breakdown
rem
prompt THE SGA AREA ALLOCATION:
prompt
prompt
prompt This shows the allocation of SGA storage. Examine this before and
prompt after making changes in the INIT.ORA file which will impact the SGA.
prompt
col name format a40
select name, bytes
from v$sgastat
set head off
select 'total of SGA '||sum(bytes)
from v$sgastat
prompt
prompt------------------------------------------------------------------------
rem
rem Displays all the base session statistics
rem
set head on
set pagesize 110
column name format a55 heading 'Statistic Name'
column value format 9,999,999,999 heading 'Result'
column statistic# format 9999 heading 'Stat#'
ttitle center 'Instance Statistics' skip 2
prompt
prompt Below is a dump of the core Instance Statistics that are greater than0.
prompt Although there are a great many statistics listed, the ones of greatest
prompt value are displayed in other formats throughout this report. Of
prompt interest here are the values for:
prompt
prompt cumulative logons
prompt(# of actual connections to the DB since last startup - good
prompt volume-of-use statistic)
prompt
prompt #93 table fetch continued row
prompt (# of chained rows - will be higher if there are a lot of long fields
prompt if the value goes up over time, it is a good signaller of general
prompt database fragmentation)
prompt
select statistic#, name, value
from v$sysstat
where value > 0
prompt
prompt -----------------------------------------------------------------------
set pages 66;
set space 3;
set heading on;
prompt
prompt Parse Ratio usually falls between 1.15 and 1.45. If it is higher, then
prompt it is usually a sign of poorly written Pro* programs or unoptimized
prompt SQL*Forms applications.
prompt
prompt Recursive Call Ratio will usually be between
prompt
prompt 7.0 - 10.0 for tuned production systems
prompt 10.0 - 14.5 for tuned development systems
prompt
prompt Buffer Hit Ratio is dependent upon RDBMS size, SGA size and
prompt the types of applications being processed. This shows the %-age
prompt of logical reads from the SGA as opposed to total reads - the
prompt figure should be as high as possible. The hit ratio can be raised
prompt by increasing DB_BUFFERS, which increases SGA size. By turning on
prompt the "Virtual Buffer Manager" (db_block_lru_statistics = TRUE and
prompt db_block_lru_extended_statistics = TRUE in the init.ora parameters),
prompt you can determine how many extra hits you would get from memory as
prompt opposed to physical I/O from disk. **NOTE: Turning these on will
prompt impact performance. One shift of statistics gathering should be enough
prompt to get the required information.
prompt
ttitle left 'Ratios for this Instance' skip 2
column pcc heading 'Parse|Ratio' format 99.99
column rcc heading 'Recsv|Cursr' format 99.99
column hr heading 'Buffer|Ratio' format 999,999,999.999
column rwr heading 'Rd/Wr|Ratio' format 999,999.9
column bpfts heading 'Blks per|Full TS' format 999,999
REM Modified for O7.1 to reverse 'cumulative opened cursors' to
REM 'opened cursors cumulative'
REM was:sum(decode(a.name,'cumulative opened cursors',value, .00000000001))
REM pcc,
REM and:sum(decode(a.name,'cumulative opened cursors',value,.00000000001))
REM rcc,
select sum(decode(a.name,'parse count',value,0)) /
sum(decode(a.name,'opened cursors cumulative',value,.00000000001)) pcc,
sum(decode(a.name,'recursive calls',value,0)) /
sum(decode(a.name,'opened cursors cumulative',value,.00000000001)) rcc,
(1-(sum(decode(a.name,'physical reads',value,0)) /
sum(decode(a.name,'db block gets',value,.00000000001)) +
sum(decode(a.name,'consistent gets',value,0))) * (-1)) hr,
sum(decode(a.name,'physical reads',value,0)) /
sum(decode(a.name,'physical writes',value,.00000000001)) rwr,
(sum(decode(a.name,'table scan blocks gotten',value,0)) -
sum(decode(a.name,'table scans (short tables)',value,0)) * 4) /
sum(decode(a.name,'table scans (long tables)',value,.00000000001))
bpfts
from v$sysstat a
prompt
prompt -----------------------------------------------------------------
prompt This looks at overall i/o activity against individual
prompt files within a tablespace
prompt
prompt Look for a mismatch across disk drives in terms of I/O
prompt
prompt Also, examine the Blocks per Read Ratio for heavily accessed
prompt TSs - if this value is significantly above 1 then you may have
prompt full tablescans occurring (with multi-block I/O)
prompt
prompt If activity on the files is unbalanced, move files around to balance
prompt the load. Should see an approximately even set of numbers across files
prompt
set pagesize 100;
set space 1
column pbr format 99999999 heading 'Physical|Blk Read'
column pbw format 999999 heading 'Physical|Blks Wrtn'
column pyr format 999999 heading 'Physical|Reads'
column readtim format 99999999 heading 'Read|Time'
column name format a40 heading 'DataFile Name'
column writetim format 99999999 heading 'Write|Time'
ttitle center 'Tablespace Report' skip 2
compute sum of f.phyblkrd, f.phyblkwrt on report
select fs.name name, f.phyblkrd pbr, f.phyblkwrt pbw,
f.readtim, f.writetim
from v$filestat f, v$datafile fs
where f.file# = fs.file#
order by fs.name
prompt
prompt -----------------------------------------------------------------
prompt GENERATING WAIT STATISTICS:
prompt
prompt This will show wait stats for certain kernel instances. This
prompt may show the need for additional rbs, wait lists, db_buffers
prompt
ttitle center 'Wait Statistics for the Instance' skip 2
column class heading 'Class Type'
column count heading 'Times Waited' format 99,999,999
column time heading 'Total Times' format 99,999,999
select class, count, time
from v$waitstat
where count > 0
order by class
prompt
prompt Look at the wait statistics generated above (if any). They will
prompt tell you where there is contention in the system. There will
prompt usually be some contention in any system - but if the ratio of
prompt waits for a particular operation starts to rise, you may need to
prompt add additional resource, such as more database buffers, log buffers,
prompt or rollback segments
prompt
prompt -----------------------------------------------------------------
prompt ROLLBACK STATISTICS:
prompt
ttitle off;
set linesize 80
column extents format 999 heading 'Extents'
column rssize format 999,999,999 heading 'Size in|Bytes'
column optsize format 999,999,999 heading 'Optimal|Size'
column hwmsize format 99,999,999 heading 'High Water|Mark'
column shrinks format 9,999 heading 'Num of|Shrinks'
column wraps format 9,999 heading 'Num of|Wraps'
column extends format 999,999 heading 'Num of|Extends'
column aveactive format 999,999,999 heading 'Average size|Active Extents'
column rownum noprint
select rssize, optsize, hwmsize,
shrinks, wraps, extends, aveactive
from v$rollstat
order by rownum
prompt
prompt -----------------------------------------------------------------
set linesize 80
break on report
compute sum of gets waits writes on report
ttitle center 'Rollback Statistics' skip 2
select rownum, extents, rssize,
xacts, gets, waits, writes
from v$rollstat
order by rownum
ttitle off
set heading off
prompt
prompt -----------------------------------------------------------------
prompt
prompt SORT AREA SIZE VALUES:
prompt
prompt To make best use of sort memory, the initial extent of your Users
prompt sort-work Tablespace should be sufficient to hold at least one sort
prompt run from memory to reduce dynamic space allocation. If you are getting
prompt a high ratio of disk sorts as opposed to memory sorts, setting
prompt sort_area_retained_size = 0 in init.ora will force the sort area to be
prompt released immediately after a sort finishes.
prompt
column value format 999,999,999
select 'INIT.ORA sort_area_size: '||value
from v$parameter
where name like 'sort_area_size'
select a.name, value
from v$statname a, v$sysstat
where a.statistic# = v$sysstat.statistic#
and a.name in ('sorts (disk)', 'sorts (memory)', 'sorts (rows)')
prompt
prompt -----------------------------------------------------------------
set heading on
set space 2
prompt
prompt This looks at Tablespace Sizing - Total bytes and free bytes
prompt
ttitle center 'Tablespace Sizing Information' Skip 2
column tablespace_name format a30 heading 'TS Name'
column sbytes format 9,999,999,999 heading 'Total Bytes'
column fbytes format 9,999,999,999 heading 'Free Bytes'
column kount format 999 heading 'Ext'
compute sum of fbytes on tablespace_name
compute sum of sbytes on tablespace_name
compute sum of sbytes on report
compute sum of fbytes on report
break on report
select a.tablespace_name, a.bytes sbytes,
sum(b.bytes) fbytes, count(*) kount
from dba_data_files a, dba_free_space b
where a.file_id = b.file_id
group by a.tablespace_name, a.bytes
order by a.tablespace_name
set linesize 80
prompt
prompt A large number of Free Chunks indicates that the tablespace may need
prompt to be defragmented and compressed.
prompt
prompt -----------------------------------------------------------------
set heading off
ttitle off
column value format 99,999,999,999
select 'Total Physical Reads', value
from v$sysstat
where statistic# = 39
prompt
prompt If you can significantly reduce physical reads by adding incremental
prompt data buffers...do it. To determine whether adding data buffers will
prompt help, set db_block_lru_statistics = TRUE and
prompt db_block_lru_extended_statistics = TRUE in the init.ora parameters.
prompt You can determine how many extra hits you would get from memory as
prompt opposed to physical I/O from disk. **NOTE: Turning these on will
prompt impact performance. One shift of statistics gathering should be enough
prompt to get the required information.
prompt
set heading on
clear computes
ttitle off
prompt
prompt -----------------------------------------------------------------
prompt CHECKING FOR FRAGMENTED DATABASE OBJECTS:
prompt
prompt Fragmentation report - If number of extents is approaching Maxextents,
prompt it is time to defragment the table.
prompt
column owner noprint new_value owner_var
column segment_name format a30 heading 'Object Name'
column segment_type format a9 heading 'Table/Indx'
column sum(bytes) format 999,999,999 heading 'Bytes Used'
column count(*) format 999 heading 'No.'
break on owner skip page 2
ttitle center 'Table Fragmentation Report' skip 2 -
left 'creator: ' owner_var skip 2
select a.owner, segment_name, segment_type,
sum(bytes), max_extents, count(*)
from dba_extents a, dba_tables b
where segment_name = b.table_name
having count(*) > 3
group by a.owner, segment_name, segment_type, max_extents
order by a.owner, segment_name, segment_type, max_extents
ttitle center 'Index Fragmentation Report' skip 2 -
left 'creator: ' owner_var skip 2
select a.owner, segment_name, segment_type,
sum(bytes), max_extents, count(*)
from dba_extents a, dba_indexes b
where segment_name = index_name
having count(*) > 3
group by a.owner, segment_name, segment_type, max_extents
order by a.owner, segment_name, segment_type, max_extents
prompt
prompt -----------------------------------------------------------------
spool off
Adith