View with billion data

Question

DEAR EXPERTS,
TABLE_A WITH 25000 REC
TABLE_B WITH 300000 REC
THE SELECT STATEMENT USED IN VIEW AND DATASET WILL RETURN 25000 X 2300 ie. one RESULT SET IS
TABLE_A WILL GET 2300 MATCHING ROWS IN TABLE_B
I NEED TOP 100 REC FROM EACH RESULTSET FROM 25000 SETS
I AM TRYING WITH THIS 2 APPROCHES.. SUGGEST ME WHICH IS BEST OR ANY OTHER WAY TO GET MY RESULT
INSERT INTO TABLE_C
WITH DATASET
AS
SELECT A.CID,
     A.ZIPCD,
     ROW_NUMBER() OVER (PARTITION BY A.CID,A.ZIPCD ORDER BY GET_DIST_FN(A.LAT,B.LAT,A.LONG,B.LONG) RANK
     FROM TABLE_A A,TABLE_B B
     WHERE A.CID = B.CID
SELECT * FROM DATASET WHERE RANK < 100
CREATE VIEW RES_VW AS
     SELECT A.CID,
          A.ZIPCD,
          ROW_NUMBER() OVER (PARTITION BY A.CID,A.ZIPCD ORDER BY GET_DIST_FN(A.LAT,B.LAT,A.LONG,B.LONG) RANK
     FROM TABLE_A A,TABLE_B B
     WHERE A.CID = B.CID
INSERT INTO TABLE_C
SELECT * FROM RES_VW WHERE RANK < 100
MANY THANKS
KALINGA

Answer 1

WE'VE ONLY HAD 4 HRS 25 MINS TO READ YOUR SHOUTING WE HAVN'T HAD TIME TO DIGEST IT YET BOTH SOLUTIONS LOOK THE SAME TO ME