请教一个问题，如此数据冗余如何删除
数据表：
CREATE TABLE tb_Table
(
ID int IDENTITY(1,1) NOT NULL,
GeoNO varchar(50) NOT NULL,
Title varchar(100) NULL,
ConID varchar(50) NULL
)

Insert into tb_Table (GeoNo,Title,ConID) values ('01','北京','1000')
Insert into tb_Table (GeoNo,Title,ConID) values ('0101','海淀区','1000')
Insert into tb_Table (GeoNo,Title,ConID) values ('010101','上地','1000')

Insert into tb_Table (GeoNo,Title,ConID) values ('01','北京','1001')
Insert into tb_Table (GeoNo,Title,ConID) values ('0102','朝阳区','1001')
Insert into tb_Table (GeoNo,Title,ConID) values ('010201','北辰','1001')

Insert into tb_Table (GeoNo,Title,ConID) values ('01','北京','1002')
Insert into tb_Table (GeoNo,Title,ConID) values ('0103','昌平区','1002')
Insert into tb_Table (GeoNo,Title,ConID) values ('010301','回龙观','1002')

我想把上述数据入库后，再从表里删除数据，保留的数据是：
'010101','上地','1000'
'010201','北辰','1001'
'010301','回龙观','1002'
其他的数据则删除，即保证ConID是唯一的，而且GeoNo的值是最大。如何用SQL来实现？
谢谢！


------解决方案--------------------

SQL code

DELETE FROM T1
FROM TB_TABLE T1
WHERE EXISTS(
SELECT 1 FROM TB_TABLE T2 WHERE T2.CONID=T1.CONID AND LEN(T2.GEOID)>LEN(T1.GEOID)
)

------解决方案--------------------
SQL code
delete t from  tb_Table t 
where exists( select 1 from  tb_Table where ConID=t.ConID and GeoNo>t.GeoNo)

------解决方案--------------------
SQL code

DELETE FROM   TB_TABLE T1
WHERE EXISTS(
SELECT 1 FROM (select max(GEOID) as GEOID,CONID from TB_TABLE group by CONID) T2  WHERE T2.CONID=T1.CONID AND  T2.GEOID>T1.GEOID)

------解决方案--------------------
SQL code
--这样似乎更合理点。
delete t from  tb_Table t 
where exists( select 1 from  tb_Table where ConID=t.ConID and len(GeoNo)>len(t.GeoNo))

------解决方案--------------------
SQL code
DELETE FROM T1
FROM TB_TABLE T1
WHERE EXISTS(
SELECT 1 FROM TB_TABLE T2 WHERE T2.CONID=T1.CONID AND LEN(T2.GEOID)>LEN(T1.GEOID)
)