如何删除仅仅第一列相同的记录？

问题描述：关系数据库中手工输入表InputData中有如下记录，

TagName	TimeStamp	Value
MM0008	2007-08-12 0:00:00	1000
MM0008	2007-08-13 0:00:00	1001
MM0008	2007-08-14 0:00:00	1002
MM0008	2007-08-15 0:00:00	1003
MM0008	2007-08-16 0:00:00	1004
MM0008	2007-08-17 0:00:00	1005
MM0008	2007-09-07 9:50:22	4444
MM0008	2007-09-07 15:10:56	100
MM0008	2007-09-07 15:11:56	200
MM0008	2007-09-11 15:53:13	500
MM0008	2007-09-11 15:54:51	500
MM0008	2007-09-11 15:55:03	123
MM0008	2007-09-11 15:58:08	500
MM0008	2007-09-11 16:01:58	0
MM0009	2007-09-11 15:58:08	501
MM0009	2007-09-11 16:01:58	0
MM0010	2007-09-11 15:58:08	502
MM0010	2007-09-11 16:01:58	0
MM0011	2007-09-11 15:58:08	503
MM0011	2007-09-11 16:01:58	0
MM0018	2007-09-11 15:58:08	504
MM0018	2007-09-11 16:01:58	0
MM0019	2007-09-11 15:58:08	505
MM0019	2007-09-11 16:01:58	0
MM0020	2007-09-11 15:58:08	506
MM0020	2007-09-11 16:01:58	0
MM0021	2007-09-11 15:58:08	507
MM0021	2007-09-11 16:01:58	0
MM0022	2007-09-11 15:58:08	508
MM0022	2007-09-11 16:01:58	0
MM0023	2007-09-11 15:58:08	509
MM0023	2007-09-11 16:01:58	0
MM0024	2007-09-11 15:58:08	510
MM0024	2007-09-11 16:01:58	0
MM0025	2007-09-11 15:58:08	511
MM0025	2007-09-11 16:01:58	0

如何获得每个TagName离当前时间最近的一条记录，把它的值及其时间插入到实时数据库中？最后得到的记录集如下所示：

TagName

TimeStamp

Value

MM00082007-09-11 16:01:580
MM00092007-09-11 16:01:580
MM00102007-09-11 16:01:580
MM00112007-09-11 16:01:580
MM00182007-09-11 16:01:580
MM00192007-09-11 16:01:580
MM00202007-09-11 16:01:580
MM00212007-09-11 16:01:580
MM00222007-09-11 16:01:580
MM00232007-09-11 16:01:580
MM00242007-09-11 16:01:580
MM00252007-09-11 16:01:580

解决方案：

1. 使用distinct：可以删除完全相同的记录，但是我现在只需要删除TagName列相同的记录，不是很好实现。

2. 使用group by

SELECT Tagname, TimeStamp, Value
FROM InputData
WHERE (TimeStamp IN
(SELECT MAX(TimeStamp) AS TimeStamp
FROM InputData
GROUP BY Tagname))