# Map bounding box.. **URL:** \ **Category:** Other MySQL® Questions\ **Created:** [January 1, 2007, 1:20pm UTC](https://forums.percona.com/t/map-bounding-box/157 "2007-01-01T13:20:11Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![oordopjes](https://avatars.discourse-cdn.com/v4/letter/o/958977/32.png) [@oordopjes](https://forums.percona.com/u/oordopjes)\ **Post date:** [January 1, 2007, 1:20pm UTC](https://forums.percona.com/t/map-bounding-box/157/1 "2007-01-01T13:20:11Z")
Hi, I would really like any suggestions on the following concerning performance…: I have 1.000.000 locations with: - a name - a latitude - a longitude - a score I want to: - select all locations on a particular (google)map. So a latitude BETWEEN (… AND …) AND longitude BETWEEN (… AND …) - fulltext match name. So name MATCHES (…) - ORDER BY score How would I get the best performance? What indexes should I create? Should I split the data in two tables and then JOIN? Or should I put everything together in one table? ANY help would be great!! ---
**Author:** ![Peter](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/peter/32/2_2.png) [@Peter](https://forums.percona.com/u/Peter)\ **Post date:** [January 1, 2007, 2:17pm UTC](https://forums.percona.com/t/map-bounding-box/157/2 "2007-01-01T14:17:37Z")
1Million is not too much, depending on your traffic of course, you might simply be able to use MySQL Full Text Search plus extra filtering by distance. If there are a lot of matches and they are filtered by distance match you’re stuck however. Sphinx could be patched to do what you need to do really fast. ---
**Author:** ![oordopjes](https://avatars.discourse-cdn.com/v4/letter/o/958977/32.png) [@oordopjes](https://forums.percona.com/u/oordopjes)\ **Post date:** [January 1, 2007, 5:40pm UTC](https://forums.percona.com/t/map-bounding-box/157/3 "2007-01-01T17:40:28Z")
I did not know of Sphynx… it seems as a good (extra) solution to my problem. I want to do something similar to the search on [yelp.com](http://yelp.com) or as is specified here. At this time I’m using MySQL 4.1 and the following query: SELECT SQL\_CALC\_FOUND\_ROWS entries.\* FROM entries JOIN geolocations ON (entries.id = geolocations.entry\_id) WHERE ( geolocations.wgs84\_lat BETWEEN 55.9920055076675 AND 57.0005291105997 ) AND ( geolocations.wgs84\_lon BETWEEN 3.55064105987549 AND 3.57124042510986 ) AND MATCH (name) AGAINST (‘+name\*’ IN BOOLEAN MODE) ORDER BY score DESC LIMIT 0, 8; I also tried putting lat and lng in the entries table…but that was not helping very much… Great forum BTW! Seems to be much more active than the original MySQL forum… ---
**Author:** ![bluesaga](https://avatars.discourse-cdn.com/v4/letter/b/f19dbf/32.png) [@bluesaga](https://forums.percona.com/u/bluesaga)\ **Post date:** [January 4, 2007, 5:34am UTC](https://forums.percona.com/t/map-bounding-box/157/4 "2007-01-04T05:34:49Z")
try not using “SQL\_FOUND\_CALC\_ROWS” a friend of mine was using this with distance and it made the queries take a LOT longer than expected. ---
**Author:** ![oordopjes](https://avatars.discourse-cdn.com/v4/letter/o/958977/32.png) [@oordopjes](https://forums.percona.com/u/oordopjes)\ **Post date:** [January 4, 2007, 5:48am UTC](https://forums.percona.com/t/map-bounding-box/157/5 "2007-01-04T05:48:48Z")
Does Sphinx allow you to use more FULLTEXT indexes? Or is it the same as with MySQL? ---
**Author:** ![bluesaga](https://avatars.discourse-cdn.com/v4/letter/b/f19dbf/32.png) [@bluesaga](https://forums.percona.com/u/bluesaga)\ **Post date:** [January 4, 2007, 5:54am UTC](https://forums.percona.com/t/map-bounding-box/157/6 "2007-01-04T05:54:44Z")
it allows you to use more than one index at a time. However it might not be how your thinking it would 😉 ---
**Author:** ![Peter](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/peter/32/2_2.png) [@Peter](https://forums.percona.com/u/Peter)\ **Post date:** [January 4, 2007, 6:28am UTC](https://forums.percona.com/t/map-bounding-box/157/7 "2007-01-04T06:28:31Z")
With Sphinx you can build as many indexes as you want and also you can search only some of the columns. It is however External search system ---
**Author:** ![oordopjes](https://avatars.discourse-cdn.com/v4/letter/o/958977/32.png) [@oordopjes](https://forums.percona.com/u/oordopjes)\ **Post date:** [January 4, 2007, 6:44am UTC](https://forums.percona.com/t/map-bounding-box/157/8 "2007-01-04T06:44:15Z")
I was thinking of doing something like: - MATCH (colx, coly) AGAINST (‘+myname\* +moreinfo\*’ IN BOOLEAN MODE) (But this is als supported by MySQL I just found out!) - MATCH (colx) AGAINST (‘this’) AND MATCH (coly) AGAINST (‘+that\*’ IN BOOLEAN MODE) Or something… I read everywhere that Sphinx is much better (read faster) in fulltext matching. Does it also support boolean matches?