Redo log 是用於恢復和一個高級特性的重要數據
一個redo條目包含了相應操作導致的數據庫變化的所有信息
所有redo條目最終都要被寫入redo文件中去
Redo log buffer是為了避免Redo文件IO導致性能瓶頸而在sga中分配出的一塊內存
一個redo條目首先在用戶內存(PGA)中產生
然後由oracle服務進程拷貝到log buffer中
當滿足一定條件時
再由LGWR進程寫入redo文件
由於log buffer是一塊
共享
內存
為了避免沖突
它是受到redo allocation latch保護的
每個服務進程需要先獲取到該latch才能分配redo buffer
因此在高並發且數據修改頻繁的oltp系統中
我們通常可以觀察到redo allocation latch的等待
Redo寫入redo buffer的整個過程如下
在PGA中生產Redo Enrey > 服務進程獲取Redo Copy latch(存在多個CPU_COUNT*) > 服務進程獲取redo allocation latch(僅個) > 分配log buffer > 釋放redo allocation latch > 將Redo Entry寫入Log Buffer > 釋放Redo Copy latch;
shared strand
為了減少redo allocation latch等待在oracle 中引入了log buffer的並行機制其基本原理就是將log buffer劃分為多個小的buffer這些小的buffer被成為strand(為了和之後出現的private strand區別它們被稱之為shared strand)每一個strand受到一個單獨redo allocation latch的保護多個shared strand的出現使原來序列化的redo buffer分配變成了並行的過程從而減少了redo allocation latch等待
shared strand的初始數據量是由參數log_parallelism控制的;在g中該參數成為隱含參數並新增參數_log_parallelism_max控制shared strand的最大數量;_log_parallelism_dynamic則控制是否允許shared strand數量在_log_parallelism和_log_parallelism_max之間動態變化
HELLODBA
COM>select nam
ksppinm
val
KSPPSTVL
nam
ksppdesc
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm LIKE
_%
AND upper(nam
ksppinm) LIKE
%LOG_PARALLE%
;
KSPPINM KSPPSTVL KSPPDESC
_log_parallelism
Number of log buffer strands
_log_parallelism_max
Maximum number of log buffer strands
_log_parallelism_dynamic TRUE Enable dynamic strands
每一個shared strand的大小 = log_buffer/(shared strand數量)strand信息可以由表x$kcrfstrand查到(包含shared strand和後面介紹的private strandg以後存在)
HELLODBA
COM>select indx
strand_size_kcrfa from x$kcrfstrand where last_buf_kcrfa !=
;
INDX STRAND_SIZE_KCRFA
HELLODBA
COM>show parameter log_buffer
NAME TYPE VALUE
log_buffer integer
關於shared strand的數量設置個cpu之內最大默認為當系統中存在redo allocation latch等待時每增加個cpu可以考慮增加個strand最大不應該超過並且_log_parallelism_max不允許大於cpu_count
注意在g中參數_log_parallelism被取消shared strand數量由_log_parallelism_max_log_parallelism_dynamic和cpu_count控制
Private strand
為了進一步降低redo buffer沖突在g中引入了新的strand機制——Private strandPrivate strand不是從log buffer中劃分的而是在shared pool中分配的一塊內存空間
HELLODBA
COM>select * from V$sgastat where name like
%strand%
;
POOL NAME BYTES
shared pool private strands
HELLODBA
COM>select indx
strand_size_kcrfa from x$kcrfstrand where last_buf_kcrfa =
;
INDX STRAND_SIZE_KCRFA
Private strand的引入為Oracle的Redo/Undo機制帶來很大的變化每一個Private strand受到一個單獨的redo allocation latch保護每個Private strand作為私有的strand只會服務於一個活動事務獲取到了Private strand的用戶事務不是在PGA中而是在Private strand生成Redo當flush private strand或者commit時Private strand被批量寫入log文件中如果新事務申請不到Private strand的redo allocation latch則會繼續遵循舊的redo buffer機制申請寫入shared strand中事務是否使用Private strand可以由x$ktcxb的字段ktcxbflg的新增的第位鑒定
HELLODBA
COM>select decode(bitand(ktcxbflg
)
) used_private_strand
count(*)
from x$ktcxb
where bitand(ksspaflg
) !=
and bitand(ktcxbflg
) !=
group by bitand(ktcxbflg
);
USED_PRIVATE_STRAND COUNT(*)
對於使用Private strand的事務無需先申請Redo Copy Latch也無需申請Shared Strand的redo allocation latch而是flush或commit是批量寫入磁盤因此減少了Redo Copy Latch和redo allocation latch申請/釋放次數也減少了這些latch的等待從而降低了CPU的負荷過程如下
事務開始 > 申請Private strand的redo allocation latch (申請失敗則申請Shared Strand的redo allocation latch) > 在Private strand中生產Redo Enrey > Flush/Commit > 申請Redo Copy Latch > 服務進程將Redo Entry批量寫入Log File > 釋放Redo Copy Latch > 釋放Private strand的redo allocation latch
注意對於未能獲取到Private strand的redo allocation latch的事務在事務結束前即使已經有其它事務釋放了Private strand也不會再申請Private strand了
每個Private strand的大小為Kg中shared pool中的Private strands的大小就是活躍會話數乘以K而g中在shared pool中需要為每個Private strand額外分配k的管理空間即數量*k
g:
SQL> select * from V$sgastat where name like
%strand%
;
POOL NAME BYTES
shared pool private strands
HELLODBA
COM>select trunc(value * KSPPSTVL /
) *
*
from (select value from v$parameter where name =
transactions
) a
(select val
KSPPSTVL
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm =
_log_private_parallelism_mul
) b;
TRUNC(VALUE*KSPPSTVL/
)*
*
g:
HELLODBA
COM>select * from V$sgastat where name like
%strand%
;
POOL NAME BYTES
shared pool private strands
HELLODBA
COM>select trunc(value * KSPPSTVL /
) * (
+
) *
from (select value from v$parameter where name =
transactions
) a
(select val
KSPPSTVL
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm =
_log_private_parallelism_mul
) b;
TRUNC(VALUE*KSPPSTVL/
)*(
+
)*
Private strand的數量受到個方面的影響logfile的大小和活躍事務數量
參數_log_private_mul指定了使用多少logfile空間預分配給Private strand默認為我們可以根據當前logfile的大小(要除去預分配給log buffer的空間)計算出這一約束條件下能夠預分配多少個Private strand
HELLODBA
COM>select bytes from v$log where status =
CURRENT
;
BYTES
HELLODBA
COM>select trunc(((select bytes from v$log where status =
CURRENT
)
(select to_number(value) from v$parameter where name =
log_buffer
))*
(select to_number(val
KSPPSTVL)
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm =
_log_private_mul
) /
/
)
as
calculated private strands
from dual;
calculated private strands
HELLODBA
COM>select count(
)
actual private strands
from x$kcrfstrand where last_buf_kcrfa =
;
actual private strands
當logfile切換後(和checkpoint一樣切換之前必須要將所有Private strand的內容flush到logfile中因此我們在alert log中可能會發現日志切換信息之前會有這樣的信息Private strand flush not complete這是可以被忽略的)會重新根據切換後的logfile的大小計算對Private strand的限制
HELLODBA
COM>alter system switch logfile;
System altered
HELLODBA
COM>select bytes from v$log where status =
CURRENT
;
BYTES
HELLODBA
COM>select trunc(((select bytes from v$log where status =
CURRENT
)
(select to_number(value) from v$parameter where name =
log_buffer
))*
(select to_number(val
KSPPSTVL)
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm =
_log_private_mul
) /
/
)
as
calculated private strands
from dual;
calculated private strands
HELLODBA
COM>select count(
)
actual private strands
from x$kcrfstrand where last_buf_kcrfa =
;
actual private strands
參數_log_private_parallelism_mul用於推算活躍事務數量在最大事務數量中的百分比默認為Private strand的數量不能大於活躍事務的數量
HELLODBA
COM>show parameter transactions
NAME TYPE VALUE
transactions integer
transactions_per_rollback_segment integer
HELLODBA
COM>select trunc((select to_number(value) from v$parameter where name =
transactions
) *
(select to_number(val
KSPPSTVL)
from sys
x$ksppi nam
sys
x$ksppsv val
where nam
indx = val
indx
AND nam
ksppinm =
_log_private_parallelism_mul
) /
)
as
calculated private strands
from dual;
calculated private strands
HELLODBA
COM>select count(
)
actual private strands
from x$kcrfstrand where last_buf_kcrfa =
;
actual private strands
注在預分配Private strand時會選擇上述個條件限制下最小一個數量但相應的shared pool的內存分配和redo allocation latch的數量是按照活躍事務數預分配的
因此如果logfile足夠大_log_private_parallelism_mul與實際活躍進程百分比基本相符的話Private strand的引入基本可以消除redo allocation latch的爭用問題
From:http://tw.wingwit.com/Article/program/Oracle/201311/17848.html