oracle数据入库延时,数据库入库慢，有等待事件，导致入库延迟！

Cache Sizes

Begin End

Buffer Cache: 4,928M 4,864M Std Block Size: 8K

Shared Pool Size: 1,792M 1,856M Log Buffer: 161,716K

Load Profile

Per Second Per Transaction Per Exec Per Call

DB Time(s): 4.5 1.8 0.09 0.07

DB CPU(s): 0.5 0.2 0.01 0.01

Redo size: 1,227,285.5 487,831.3

Logical reads: 8,587.5 3,413.4

Block changes: 5,057.9 2,010.5

Physical reads: 152.9 60.8

Physical writes: 248.9 98.9

User calls: 60.5 24.1

Parses: 44.8 17.8

Hard parses: 2.5 1.0

W/A MB processed: 3,194,233.6 1,269,669.6

Logons: 5.7 2.3

Executes: 51.3 20.4

Rollbacks: 0.1 0.0

Transactions: 2.5

Instance Efficiency Percentages (Target 100%)

Buffer Nowait %: 99.08 Redo NoWait %: 100.00

Buffer Hit %: 99.23 In-memory Sort %: 100.00

Library Hit %: 94.67 Soft Parse %: 94.50

Execute to Parse %: 12.76 Latch Hit %: 99.87

Parse CPU to Parse Elapsd %: 0.00 % Non-Parse CPU: 88.92

Shared Pool Statistics

Begin End

Memory Usage %: 88.22 82.63

% SQL with executions>1: 73.91 71.71

% Memory for SQL w/exec>1: 76.90 76.76

Top 5 Timed Foreground Events

Event Waits Time(s) Avg wait (ms) % DB time Wait Class

enq: TM - contention 331 4,048 12230 24.77 Application

SQL*Net message from dblink 33,549 3,181 95 19.47 Network

DB CPU   1,923   11.77

log file sync 12,053 1,712 142 10.48 Commit

row cache lock 340 1,138 3348 6.97 Concurrency

这事我抓的最近一个小时awr报告，里面有等待事件。

enq: TM - contention

对于log file sync等待事件，目前我们是批量入库的批量提交的，并且是写到raid0上了，之前没有任何问题，现在有这个问题，主要是还是受入库enq: TM - contention 这个事件的影响。

row cache lock 这个等待事件，我还没有找到，不知道怎么定位。