oracle分区表扩分区 很慢,升级oracle10g 大分区表update变慢

原来使用的数据库是oracle8.1.7.4   这次升级为oracle 10.0.2.4,在导入旧dmp时候明显能够感觉到导入数据比较快。

1.可是在正式使用之后，发现启用一个大表(使用了分区表)，在update使用比较慢，语句和索引建的和过去在oracle8i上是完全一致的。这是什么原因？也重建过索引，还是有这个问题。

2.在使用db_links从该数据库向一个oracle8i传递数据时候，很简单的过程，就是这边读然后写入另一个数据库，速度也特别慢，这是怎么回事？

请高手指教。

3.另：alert.log出现如下

Sat Aug  4 22:46:37 2012

Thread 1 cannot allocate new log, sequence 819

Checkpoint not complete

Current log# 2 seq# 818 mem# 0: /u01/oradata/jf68/redo02.log

Sat Aug  4 22:46:41 2012

Thread 1 advanced to log sequence 819 (LGWR switch)

Current log# 3 seq# 819 mem# 0: /u01/oradata/jf68/redo03.log

Sat Aug  4 22:47:04 2012

Thread 1 cannot allocate new log, sequence 820

Checkpoint not complete

Current log# 3 seq# 819 mem# 0: /u01/oradata/jf68/redo03.log

Sat Aug  4 22:47:06 2012

Thread 1 advanced to log sequence 820 (LGWR switch)

Current log# 1 seq# 820 mem# 0: /u01/oradata/jf68/redo01.log

Sat Aug  4 22:48:00 2012

Thread 1 cannot allocate new log, sequence 821

Checkpoint not complete

Current log# 1 seq# 820 mem# 0: /u01/oradata/jf68/redo01.log

Sat Aug  4 22:48:08 2012

Thread 1 advanced to log sequence 821 (LGWR switch)

Current log# 2 seq# 821 mem# 0: /u01/oradata/jf68/redo02.log

Sun Aug  5 07:27:12 2012

Thread 1 advanced to log sequence 822 (LGWR switch)

Current log# 3 seq# 822 mem# 0: /u01/oradata/jf68/redo03.log

Sun Aug  5 07:27:50 2012

Thread 1 advanced to log sequence 823 (LGWR switch)

Current log# 1 seq# 823 mem# 0: /u01/oradata/jf68/redo01.log

Sun Aug  5 07:31:20 2012

这是怎么回事？