CREATE TABLE mytable
( CId int
, CCId varchar(10)
, CCount int
, totalCount int
)
;
GO
INSERT MyTable 
VALUES (1, 'a',3, 6)
, (1, 'a', 3, 6)
, (1, 'b', 3, 6)
, (1, 'c', 3, 6)
, (2, 'b', 2, 6)
, (2, 'b', 2, 6)
, (2, 'a', 2, 6)
, (2, 'a', 2, 6)
, (3, 'v', 1, 6)
;
go

SELECT cid, COUNT(DISTINCT ccid)
 FROM dbo.MyTable
 GROUP BY cid

;WITH SampleData AS (
	SELECT CId = 1, CCId = 'a', CCount = 3, totalCount = 6 UNION ALL
	SELECT 1, 'a', 3, 6 UNION ALL
	SELECT 1, 'b', 3, 6 UNION ALL
	SELECT 1, 'c', 3, 6 UNION ALL
	SELECT 2, 'b', 2, 6 UNION ALL
	SELECT 2, 'b', 2, 6 UNION ALL
	SELECT 2, 'a', 2, 6 UNION ALL
	SELECT 2, 'a', 2, 6 UNION ALL
	SELECT 3, 'v', 1, 6
) 

SELECT 
	CId, 
	CCId, 
	CCount, 
	totalCount = SUM([First]) OVER (PARTITION BY (SELECT NULL))
FROM (
	SELECT *, 
		[First] = CASE WHEN 1 = ROW_NUMBER() OVER(PARTITION BY CId ORDER BY CCId) THEN CCount END
	FROM SampleData
) d