只要没有多个相同的c，就可以使用

select t1.* 
from table1 t1 JOIN (SELECT b, MAX(c) max_C FROM table1) t2 
ON t1.b = t2.b AND t1.c = t2.max_c

或者如果您的数据库支持cte和窗口函数，并且您有多个相同的c

WITH CTE as    
(select t1.*, ROW_NUMBER() OVER(PARTITION BY b ORDER BY c DESC,a DESC) rn
    from table1 t1)
SELECT a,b,c FROM CTE WHERE rn = 1

对于sql-server，我创建了这个sql-fiddle
创建一个简单的表

CREATE TABLE my_table (
    A_id int, B int, C_id int
);

给它添加一些数据

-- group by b but select only record where c is max
INSERT INTO my_table (A_id, B, C_id)  VALUES (1,2,3);
INSERT INTO my_table (A_id, B, C_id)  VALUES (2,2,4); -- this c_id=4 or
INSERT INTO my_table (A_id, B, C_id)  VALUES (3,2,4); -- that c_id=4 or both?
INSERT INTO my_table (A_id, B, C_id)  VALUES (4,3,40);
INSERT INTO my_table (A_id, B, C_id)  VALUES (5,3,41); 
INSERT INTO my_table (A_id, B, C_id)  VALUES (6,3,42); -- only this 
INSERT INTO my_table (A_id, B, C_id)  VALUES (7,4,50);
INSERT INTO my_table (A_id, B, C_id)  VALUES (8,4,51);
INSERT INTO my_table (A_id, B, C_id)  VALUES (9,4,52); -- only this

上面的插入创建了这些记录

然后查询数据

SELECT my_table.A_id, sq.B, sq.C_Max FROM 
   (SELECT B, Max(C_id) as C_Max FROM my_table GROUP BY B) as sq
LEFT JOIN my_table ON sq.B = my_table.B AND sq.C_Max = my_table.C_id;

这部分SELECT B, Max(C_id) as C_Max FROM my_table GROUP BY B获取C_id具有最高值且仅包括列B和C的所有记录。此结果可用于将结果与表本身连接以查找匹配行。