Should I prefilter my sql server statement or have power query

[shophoney]

I'm looking to speed up my Excel file with lots of data being pulled from our sql server system.

I've created statements like below and would like to know if I should pre filter the sales from 2015 and prior. Rather than having power query remove the unneeded rows and columns.

If so how would i add to the statement to remove sales prior to 01/01/2015?

SELECT tblsale.sale_link,
tblsale.loc_code AS [LOC#],
tblsale.DATE AS [TRANS DATE],
tblsale.TIME,
tblsale.trans_no AS [TRANS#],
tblsale.customer_code AS [CUST#],
tblsalecommission.salesperson,
tblsaleitem.sku_no AS [SKU#],
tblsaleitem.description,
Min(tblsalediscount.disc_code) AS [DISCOUNT CODE],
tblsaleitem.qty AS [UNITS SOLD],
SUM(tblsaleitem.ext_cost) AS COST,
[unit_ext_price] - [unit_ext_best_price] AS DISCOUNT,
tblsaleitem.unit_ext_price,
tblsaleitem.unit_ext_best_price AS [NET RETAIL],
tblsaleitem.line_no,
tblsale.trans_type,
tblsale.void_code,
tblsale.suspended,
tblsaleitem.affect_inv,
tblsaleitem.affect_total,
tblsale.note AS [SALE NOTE],
tblsaleitem.note AS [SALE ITEM NOTE]
FROM tblsalediscount
RIGHT JOIN (tblsale
INNER JOIN (tblsaleitem
INNER JOIN tblsalecommission
ON ( tblsaleitem.line_no =
tblsalecommission.line_no )
AND ( tblsaleitem.sale_link =
tblsalecommission.sale_link ))
ON tblsale.sale_link = tblsaleitem.sale_link)
ON ( tblsalediscount.line_no = tblsaleitem.line_no )
AND ( tblsalediscount.sale_link = tblsaleitem.sale_link )
GROUP BY tblsale.sale_link,
tblsale.loc_code,
tblsale.DATE,
tblsale.TIME,
tblsale.trans_no,
tblsale.customer_code,
tblsalecommission.salesperson,
tblsaleitem.sku_no,
tblsaleitem.description,
tblsaleitem.qty,
[unit_ext_price] - [unit_ext_best_price],
tblsaleitem.unit_ext_price,
tblsaleitem.unit_ext_best_price,
tblsaleitem.line_no,
tblsale.trans_type,
tblsale.void_code,
tblsale.suspended,
tblsaleitem.affect_inv,
tblsaleitem.affect_total,
tblsale.note,
tblsaleitem.note;

 

[gazpage]

One man's lots of data, is often Power BIs not much data. How many columns /rows are we talking? 100s of millions or billions?