Data model for storing active orders

active_orders.cql

CREATE TABLE event (
id uuid,
active text,
event_time timestamp,
event text,
PRIMARY KEY ((id), active, event_time)
);

SELECT id FROM event WHERE id=<uuid> AND acrive='Y';

CREATE TABLE active_orders (
  day text, //Date to the day. If high volume go to hour or minute for better partitioning
  order_time timestamp,
  id uuid,
  event text,
  PRIMARY KEY ((day), order_time)
) WITH CLUSTERING ORDER BY (order_time DESC);

SELECT * from active_orders WHERE day='20141030' AND order_time < '201410301200';

You'll want to reverse this to avoid any issues with tombstones and just make queries more efficient. 

As a variant on the same index:

CREATE TABLE active_orders2 (
  day text, //Date to the day. If high volume go to hour or minute for better partitioning
  product text, // If looking to track all active orders for a specific product
  order_time timestamp,
  id uuid,
  event text,
  PRIMARY KEY ((day, product), order_time)
) WITH CLUSTERING ORDER BY (order_time DESC);

// All active orders for October 30
SELECT * FROM active_orders2 WHERE day='20141030' AND product = '12345'; 

// All active orders for October 30 between 9-11AM 
SELECT * FROM active_orders2 WHERE day='20141030' AND product = '12345' AND order_time >= '201410300900' AND order_time <= '201410301100';