Exercises Data Warehousing

SQL Extensions & DB Explosion Problem

Solutions

Figure 1: Schema of the Northwind database

1. Consider the Northwind database whose schema is given in Figure 1. This database contains information
of orders placed by customers. For every order the detail is given of what products were sold, for what
unit price and in what quantity. The employee that secured the order is recorded as well as the date in
which the order was inserted. For customers the city they live in etc. is recorded, and for employees their
salesdistrict. For this database, create queries to generate the following reports:
(a) Select the number of sales per category and country.
SELECT CategoryName, Country, COUNT(*) AS Count
FROM "Order Details" O, Products P, Categories C, Suppliers S
WHERE O.ProductID = P.ProductID AND P.CategoryID = C.CategoryID AND
P.SupplierID = S.SupplierID
GROUP BY CategoryName, Country
ORDER BY CategoryName, Country
(b) Select the 3 top-selling categories overall (hint: use “select top 3” construction).
SELECT CategoryName, Country, COUNT(*) AS Count
FROM "Order Details" O, Products P, Categories C, Suppliers S

1
 WHERE O.ProductID = P.ProductID AND P.CategoryID = C.CategoryID AND
P.SupplierID = S.SupplierID
GROUP BY CategoryName
ORDER BY CategoryName
(c) Produce an overview of sales by month for these categories (hint: get month and year with “month”
and “year” functions). Are there countries and product categories for which the trend over time is
increasing?
WITH Top3Categories AS (
SELECT TOP 3 C.CategoryId, CategoryName, COUNT(*) AS Count
FROM "Order Details" O, Products P, Categories C
WHERE O.ProductID = P.ProductID AND P.CategoryID = C.CategoryID
GROUP BY C.CategoryId, CategoryName
ORDER BY Count DESC )
SELECT CategoryName, Country, year(OrderDate) AS Year,
month(OrderDate) AS Month, COUNT(*) AS Count
FROM Orders O, "Order Details" OD, Products P, Top3Categories C, Suppliers S
WHERE O.OrderID = OD.OrderID AND OD.ProductID = P.ProductID AND
P.CategoryID = C.CategoryID AND P.SupplierID = S.SupplierID
GROUP BY CategoryName, Country, year(OrderDate), month(OrderDate)
ORDER BY CategoryName, Country, year(OrderDate), month(OrderDate)
(d) List total amount of sales in $ by employee and year (discount in OrderDetails is at UnitPrice level).
Which employees have an increase in sales over the three reported years?
SELECT FirstName, LastName, year(OrderDate) AS Year,
FORMAT(SUM((1-Discount)*OD.UnitPrice*Quantity),’C’, ’en-us’) AS TotalAmount
FROM Orders O, "Order Details" OD, Employees E
WHERE O.OrderID = OD.OrderID AND O.EmployeeID = E.EmployeeID
GROUP BY FirstName, + LastName, year(OrderDate)
ORDER BY FirstName, LastName, year(OrderDate)
(e) Get an individual sales report by month for employee 9 (Dodsworth) in 1997.
SELECT month(OrderDate) AS Month,
FORMAT(SUM((1-Discount)*OD.UnitPrice*Quantity),’C’, ’en-us’) AS TotalAmount
FROM Orders O, "Order Details" OD
WHERE O.OrderID = OD.OrderID AND O.EmployeeID = 9 AND year(OrderDate) = 1997
GROUP BY month(OrderDate)
ORDER BY month(OrderDate)
(f) Get a sales report by country and month.
SELECT Country, year(OrderDate) AS Year, month(OrderDate) AS Month,
FORMAT(SUM((1-Discount)*OD.UnitPrice*Quantity),’C’, ’en-us’) AS TotalAmount
FROM Orders O, "Order Details" OD, Products P, Suppliers S
WHERE O.OrderID = OD.OrderID AND OD.ProductID = P.ProductID AND
P.SupplierID = S.SupplierID
GROUP BY Country, year(OrderDate), month(OrderDate)
ORDER BY Country, year(OrderDate), month(OrderDate)
2. The sales department of a supermarket chain wants to have a system to support the strategic planning
and evaluation of promotions. To this end, they need sales information over the different stores of the
supermarket chain. For their analysis tasks they want to compute average sales and total sales, for different
products, either at product level or brand level, for different stores at different levels of granularity:
individual store, province where the store is located, and country, and for different time periods: per year,
month, quarter, semester and also by day of the week.
(a) How would you conceptually model the data needed by the sales department as a data cube? E.g.,
what are the measures, the dimensional attributes, the hierarchies, the aggregations that are needed?
Solution: The dimensions are as follows
• Product(Product, Brand, Type),
• Store(Store, Province, Country), and

2
 • Date(Month, Semester, Year, Weekday).
The measure is sales. The aggregation functions are sum (for total sales) and average (for average
sales). The hierarchies are as follows:
Product → Brand
Store → Province → Country
Date → Month → Semester → Year
→ Weekday
(b) Given the cube of (a), explain how you would construct the answers to the following queries with
the operations slice-and-dice, pivot, roll-up, and drill-down. If necessary, indicate in which cell(s) of
the constructed cube the answer can be found:
i. Give the total overall sales per store.
Solution: Cells (Store, all, all) of the original cube. Slice on Product=all and Date=all. The
measure is TotalSales.
ii. Give an overview of the average sales per month per province.
Solution: The measure is AverageSales. Slice on Product=all. Roll-up Store to Province, and
Day to Month. Represent it using a pivot-table on dimensions Store and Day.
iii. Give the subcube with only dimensions store at level province and day at level month for the
average and total sales for the period 1999 till 2005.
Solution: Slice: date must be in 1999 till 2005. Roll-up Store to Province, Date to Month.
(c) Give an SQL:1999 expression that produces the datacube (i.e., contains all aggregates of the cube
using the null value in an attribute to represent aggregation on the corresponding dimension). How
do you handle the multiple measures? The hierarchy?
We assume that the base data is stored in the following relational tables:
• Product(ProductID, Brand, Type)
• Store(StoreID, Province, Country)
• Date(Date, Weekday, Month, Semester, Year)
• Sales(ProductID, StoreID, Day, Amount)
SELECT P.ProductID, Brand, S.StoreID, Province, Country, D.Day,
Weekday, Month, Semester, Year, SUM(Amount) AS Total, AVG(amount) AS Average
FROM Product P, Store S, Day D, Sales Sa
WHERE P.ProductID = SA.ProductID and S.StoreID = Sa.StoreID and D.day = Sa.day
GROUP BY ROLLUP(brand,P.ProductID), ROLLUP(country, province, S.StoreID),
ROLLUP(year, semester, month, D.day), ROLLUP(weekday, D.day);
3. Give SQL:1999 expressions for the queries in 2(b).
Solution
Let Cube be the result of the query in 2(c).

• Give the total overall sales per store.

SELECT StoreID, Total
FROM Cube
WHERE Brand IS NULL AND Year IS NULL AND Weekday IS NULL AND StoreID IS NOT NULL
• Give an overview of the average sales per month per province.
SELECT Month, Year, Province, Average
FROM Cube
WHERE Brand IS NULL AND StoreID IS NULL AND Province IS NOT NULL AND
Date IS NULL AND Month IS NOT NULL AND Weekday IS NULL
• Give the subcube with only dimensions store at level province and date at level month for the average
and total sales for the period 1999 till 2005.
SELECT Month, Year, Province, Average, Total
FROM Cube
WHERE Brand IS NULL AND StoreID IS NULL AND Province IS NOT NULL AND
Date IS NULL AND Month IS NOT NULL AND Year >= 1999 AND Year <= 2015 AND
Weekday IS NULL

3
4. Suppose that we have a relation Sales(Product, Month, Store, Amount). There are five products: P1, P2,
P3, P4, P5, 12 months of data, and three stores: S1, S2, and S3.

(a) (Dense setting) Suppose that every product has been sold in every month in every store; i.e., for
every combination of a product p, a month m, and a store s, there is a tuple (p, m, s, a) with a
non-zero amount.
i. How many tuples does this relation contain?
Solution: 5 × 12 × 3 = 180
ii. How many tuples does a data cube with dimensions Product, Month, Store, and measure Amount
contain?
Solution: 6 × 13 × 4 = 312
(b) (Sparse setting) Consider the following (sparse) relation:
Product Month Store Amount
P1 Jan S1 a1
P1 Jan S2 a2
P2 Feb S2 a3
P2 Feb S3 a4
P3 Jan S1 a5
P3 Feb S1 a6
P4 Feb S1 a7
P5 Jan S3 a8
How many non-empty cells does the data cube of this relation contain?
Solution:
Group By # Group By #
PMS 8 P 5
PM 6 M 2
PS 7 S 3
MS 6 () 1
Hence, in total: 38 non-empty cells.