I am new here, so please be nice...
I have a question regarding TimesTen & Oracle. I have attempted an experiment with both pieces of software, and realize that the query speed on a table is slower on TimesTen than on Oracle.
The experiment setup was as follows:
- 500,000 rows of data
- count(*) query on a column which is NOT a candidate key
CREATE TABLE menu(DISH_ID NUMBER NOT NULL PRIMARY KEY, price NUMBER(7))
SELECT COUNT(*) FROM menu WHERE price <= 50. The price value increases uniformly until the largest value in the DB. Time taken for the query to be executed 1000 times was recorded.
The query speed for different ranges is consistently slower on TimesTen, as compared to Oracle. No indexes were built on the search key. Is there a particular reason for this?
Other notes: I ran the same experiment, but built an index on the search key on both TimesTen & on 10g, and the difference is stark on TimesTen's favour.