What is the ideal data type to use when storing la

Question

Bearing in mind that I'll be performing calculations on lat / long pairs, what datatype is best suited for use with a MySQL database?

Answer 1

TL;DR

Use FLOAT(8,5) if you're not working in NASA / military and not making aircrafts navi systems.

---

To answer your question fully, you'd need to consider several things:

Format

• degrees minutes seconds: 40° 26′ 46″ N 79° 58′ 56″ W
• degrees decimal minutes: 40° 26.767′ N 79° 58.933′ W
• decimal degrees 1: 40.446° N 79.982° W
• decimal degrees 2: -32.60875, 21.27812
• Some other home-made format? Noone forbids you from making your own home-centric coordinates system and store it as heading and distance from your home. This could make sense for some specific problems you're working on.

So the first part of the answer would be - you can store the coordinates in the format your application uses to avoid constant conversions back and forth and make simpler SQL queries.

Most probably you use Google Maps or OSM to display your data, and GMaps are using "decimal degrees 2" format. So it will be easier to store coordinates in the same format.

Precision

Then, you'd like to define precision you need. Of course you can store coordinates like "-32.608697550570334,21.278081997935146", but have you ever cared about millimeters while navigation to the point? If you're not working in NASA and not doing satellites or rockets or planes trajectories, you should be fine with several meters accuracy.

Commonly used format is 5 digits after dots which gives you 50cm accuracy.

Example: there is 1cm distance between X,21.2780818 and X,21.2780819. So 7 digits after dot give you 1/2cm precision and 5 digits after dot will give you 1/2 meters precision (because minimal distance between distinct points is 1m, so rounding error cannot be more than half of it). For most civil purposes it should be enough.

degrees decimal minutes format (40° 26.767′ N 79° 58.933′ W) gives you exactly the same precision as 5 digits after dot

Space-efficient storage

If you've selected decimal format, then your coordinate is a pair (-32.60875, 21.27812). Obviously, 2 x (1 bit for sign, 2 digits for degrees and 5 digits for exponent) will be enough.

So here I'd like to support Alix Axel from comments saying that Google suggestion to store it in FLOAT(10,6) is really extra, because you don't need 4 digits for main part (since sign is separated and latitude is limited to 90 and longitude is limited to 180). You can easily use FLOAT(8,5) for 1/2m precision or FLOAT(9,6) for 50/2cm precision. Or you can even store lat and long in separated types, because FLOAT(7,5) is enough for lat. See MySQL float types reference. Any of them will be like normal FLOAT and equal to 4 bytes anyway.

Usually space is not an issue nowadays, but if you want to really optimize the storage for some reason (Disclaimer: don't do pre-optimization), you may compress lat(no more than 91 000 values + sign) + long(no more than 181 000 values + sign) to 21 bits which is significantly less than 2xFLOAT (8 bytes == 64 bits)

Answer 2

depending on you application, i suggest using FLOAT(9,6)

spatial keys will give you more features, but in by production benchmarks the floats are much faster than the spatial keys. (0,01 VS 0,001 in AVG)

Answer 3

While it isn't optimal for all operations, if you are making map tiles or working with large numbers of markers (dots) with only one projection (e.g. Mercator, like Google Maps and many other slippy maps frameworks expect), I have found what I call "Vast Coordinate System" to be really, really handy. Basically, you store x and y pixel coordinates at some way-zoomed-in -- I use zoom level 23. This has several benefits:

• You do the expensive lat/lng to mercator pixel transformation once instead of every time you handle the point
• Getting the tile coordinate from a record given a zoom level takes one right shift.
• Getting the pixel coordinate from a record takes one right shift and one bitwise AND.
• The shifts are so lightweight that it is practical to do them in SQL, which means you can do a DISTINCT to return only one record per pixel location, which will cut down on the number records returned by the backend, which means less processing on the front end.

I talked about all this in a recent blog post: http://blog.webfoot.com/2013/03/12/optimizing-map-tile-generation/

Answer 4

Lat Long calculations require precision, so use some type of decimal type and make the precision at least 2 higher than the number you will store in order to perform math calculations. I don't know about the my sql datatypes but in SQL server people often use float or real instead of decimal and get into trouble because these are are estimated numbers not real ones. So just make sure the data type you use is a true decimal type and not a floating decimal type and you should be fine.

Answer 5

MySQL's Spatial Extensions are the best option because you have the full list of spatial operators and indices at your disposal. A spatial index will allow you to perform distance-based calculations very quickly. Please keep in mind that as of 6.0, the Spatial Extension is still incomplete. I am not putting down MySQL Spatial, only letting you know of the pitfalls before you get too far along on this.

If you are dealing strictly with points and only the DISTANCE function, this is fine. If you need to do any calculations with Polygons, Lines, or Buffered-Points, the spatial operators do not provide exact results unless you use the "relate" operator. See the warning at the top of 21.5.6. Relationships such as contains, within, or intersects are using the MBR, not the exact geometry shape (i.e. an Ellipse is treated like a Rectangle).

Also, the distances in MySQL Spatial are in the same units as your first geometry. This means if you're using Decimal Degrees, then your distance measurements are in Decimal Degrees. This will make it very difficult to get exact results as you get furthur from the equator.

Answer 6

When I did this for a navigation database built from ARINC424 I did a fair amount of testing and looking back at the code, I used a DECIMAL(18,12) (Actually a NUMERIC(18,12) because it was firebird).

Floats and doubles aren't as precise and may result in rounding errors which may be a very bad thing. I can't remember if I found any real data that had problems - but I'm fairly certain that the inability to store accurately in a float or a double could cause problems

The point is that when using degrees or radians we know the range of the values - and the fractional part needs the most digits.

The MySQL Spatial Extensions are a good alternative because they follow The OpenGIS Geometry Model. I didn't use them because I needed to keep my database portable.