create table #tb(aa int,bb datetime,cc varchar(50))
insert into #tb values
(65443, '2012-04-11' ,'40'), 
(65443, '2012-04-12' ,'50') , 
(65443 ,'2012-04-21' ,'250') , 
(65446 ,'2012-04-29' ,'130') ,
(65447 ,'2012-04-17' ,'test') ,
(65447 ,'2012-04-29' ,'1290'),
(65448 ,'2012-04-29' ,'650'),
(65449 ,'2012-04-28' ,'1100'),
(65449 ,'2012-04-29' ,'190');

select aa,bb,cc from
(select *,row_number()over(partition by aa order by bb desc) as rin from #tb)fin
where fin.rin=1

drop table #tb

/*
(9 row(s) affected)
aa          bb                      cc
----------- ----------------------- --------------------------------------------------
65443       2012-04-21 00:00:00.000 250
65446       2012-04-29 00:00:00.000 130
65447       2012-04-29 00:00:00.000 1290
65448       2012-04-29 00:00:00.000 650
65449       2012-04-29 00:00:00.000 190

(5 row(s) affected)
*/

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

create table tb(aa int,bb datetime,cc varchar(50))
insert into tb values
(65443, '2012-04-11' ,'40'), 
(65443, '2012-04-12' ,'50') , 
(65443 ,'2012-04-21' ,'250') , 
(65446 ,'2012-04-29' ,'130') ,
(65447 ,'2012-04-17' ,'test') ,
(65447 ,'2012-04-29' ,'1290'),
(65448 ,'2012-04-29' ,'650'),
(65449 ,'2012-04-28' ,'1100'),
(65449 ,'2012-04-29' ,'190');
select * from tb as a where not exists (select 1 from tb where aa = a.aa and bb > a.bb)
drop table tb

------解决方案--------------------
select a.aa,a.bb,b.cc from (  select aa,MAX(bb) as bb from #tb group by aa) a join #tb b on a.aa=b.aa and a.bb=b.bb