专家意见:几个办法:
　　1. 如果删除的数据是大部分,建议使用楼上的方法把要保留的数据放在一个临时表里,truncate table后再放回来
　　2. 也可以分段提交,楼上也提到了
　　3. 专门使用一个大回滚段
　　4. 如果确认将来不需要做恢复,改为非归档模式,删除完改回来再做个备份.
　　专家给出的解决方案:
有条件的分步删除数据表中的记录
--创建测试表
create table test as select * from dba_objects;
Table created.
--创建删除表的存储过程
 create or replace procedure deleteTab
--插入语句 
   SQL> insert into test select * from dba_objects;
6374 rows created.
SQL> /
6374 rows created.
SQL> /
6374 rows created.
SQL> commit;

--创建删除的存储过程
create or replace procedure deleteTab
  /**
   ** Usage: run the script to create the proc deleteTab
   **        in SQL*PLUS, type "exec deleteTab(’Foo’,’ID>=1000000’,’3000’);"
   **        to delete the records in the table "Foo", commit per 3000 records.
   **       Condition with default value ’1=1’ and default Commit batch is 10000.
   **/
  (
    p_TableName    in    varchar2,    -- The TableName which you want to delete from
    p_Condition    in    varchar2 default ’1=1’,    -- Delete condition, such as "id>=100000"
    p_Count        in    varchar2 default ’10000’    -- Commit after delete How many records
  )
  as
   pragma autonomous_transaction;
   n_delete number:=0;
  begin
   while 1=1 loop
     EXECUTE IMMEDIATE
       ’delete from ’||p_TableName||’ where ’||p_Condition||’ and rownum <= :rn’
     USING p_Count;
     if SQL%NOTFOUND then
     exit;
     else
          n_delete:=n_delete + SQL%ROWCOUNT;
     end if;
     commit;
   end loop;
   commit;
   DBMS_OUTPUT.PUT_LINE(’Finished!’); 
   DBMS_OUTPUT.PUT_LINE(’Totally ’||to_char(n_delete)||’ records deleted!’);
  end;
  /
--执行语句
SQL> exec deleteTab(’TEST’,’object_id >0’,’10000’)
你看看执行结果我试验过，效果还可以