gistfile1.txt

test=> CREATE TABLE points (id serial PRIMARY KEY, p point NOT NULL);
CREATE TABLE
test=> CREATE INDEX ON points USING GIST (p);
CREATE INDEX
test=> INSERT INTO points (p) VALUES ('1,1'), ('1,0'), ('0,1'), ('0,0.5'), ('0,2');
INSERT 0 5

test=> SELECT * FROM points ORDER BY p <-> '0,0' LIMIT 3;
 id |    p    
----+---------
  4 | (0,0.5)
  2 | (1,0)
  3 | (0,1)
(3 rows)

test=> EXPLAIN SELECT * FROM points ORDER BY p <-> '0,0' LIMIT 3;
                                      QUERY PLAN                                      
--------------------------------------------------------------------------------------
 Limit  (cost=0.14..0.28 rows=3 width=20)
   ->  Index Scan using points_p_idx on points  (cost=0.14..78.14 rows=1700 width=20)
         Order By: (p <-> '(0,0)'::point)
(3 rows)