Aggregations

YugabyteDB supports a number of standard aggregation functions. Let us go through some of these using an example. Consider a products table as shown below.

Let us create this table with ProductID as the primary hash key.

cqlsh> CREATE KEYSPACE store;

cqlsh> CREATE TABLE store.products (ProductID BIGINT PRIMARY KEY, ProductName VARCHAR, SupplierID INT, CategoryID INT, Unit TEXT, Price FLOAT, Quantity INT);

Now let us populate the sample data.

INSERT INTO store.products (ProductID, ProductName, SupplierID, CategoryID, Unit, Price, Quantity) VALUES (1, 'Chais', 1, 1, '10 boxes x 20 bags', 18, 25);
INSERT INTO store.products (ProductID, ProductName, SupplierID, CategoryID, Unit, Price, Quantity) VALUES (2, 'Chang', 1, 1, '24 - 12 oz bottles', 19, 12);
INSERT INTO store.products (ProductID, ProductName, SupplierID, CategoryID, Unit, Price, Quantity) VALUES (3, 'Aniseed Syrup', 1, 2, '12 - 550 ml bottles', 10, 10);
INSERT INTO store.products (ProductID, ProductName, SupplierID, CategoryID, Unit, Price, Quantity) VALUES (4, 'Chef Anton''s Cajun Seasoning', 2, 2, '48 - 6 oz jars', 22, 9);
INSERT INTO store.products (ProductID, ProductName, SupplierID, CategoryID, Unit, Price, Quantity) VALUES (5, 'Chef Anton''s Gumbo Mix', 2, 2, '36 boxes', 21.35, 40);

Counts

• Finding the number of item types in the store can be done as follows.

  cqlsh> SELECT COUNT(ProductID) FROM store.products;

  count(productid)
  ------------------
              5
  
  (1 rows)
  

• We can give an alias name to the count column as follows.

  cqlsh> SELECT COUNT(ProductID) as num_products FROM store.products;

  num_products
  --------------
          5
  
  (1 rows)
  

• Finding the number of item types for supplier 1 can be done as follows.

You can do this as shown below.

cqlsh> SELECT COUNT(ProductID) as supplier1_num_products FROM store.products WHERE SupplierID=1;

 supplier1_num_products
------------------------
                      3

(1 rows)

Numeric aggregation functions

The standard aggregate functions of min, max, sum, avg and count are built-in functions.

• To find the total number of items in the store, run the following query.

  cqlsh> SELECT SUM(Quantity) FROM store.products;

  sum(quantity)
  ---------------
          96
  
  (1 rows)
  

• To find the price of the cheapest and the most expensive item, run the following.

  cqlsh> SELECT MIN(Price), MAX(Price) FROM store.products;

  min(price) | max(price)
  ------------+------------
       10 |         22
  
  (1 rows)
  

• To find the average price of all the items in the store, run the following.

  cqlsh> SELECT AVG(price) FROM store.products;

  system.avg(price)
  -------------------
           18.07
  
  (1 rows)