Complex Networks in Hive - Optimization Code

Question

I have a problem with how to get my Hive code optimized. I have a huge table as follows:

Customer_id     Product_id     Date     Value
     1               1         02/28    100.0 
     1               2         02/02    120.0 
     1               3         02/10    144.0 
     2               2         02/15    120.0 
     2               3         02/28    144.0 
    ...             ...         ...     ... 

I want to create a complex network where I link the products through the buyers. The graph does not have to be directed and I have to count the number of links between them. In the end I need this:

Product_x     Product_y     amount
    1             2           1
    1             3           1
    2             3           2

Can anyone help me with this? I need an optimized way to do this. The join of the table with itself is not the solution. I really need an optimum way on this =/

CREATE TABLE X AS
SELECT 
     a.product_id as product_x,
     b.product_id as product_y,
     count(*) as amout
FROM table as a
JOIN table as b
ON a.customer_id = b.customer_id
WHERE a.product_id < b.product_id
GROUP BY product_x, product_y;