1 # Importing the Database
3 The following instructions explain how to create a Nominatim database
4 from an OSM planet file and how to keep the database up to date. It
5 is assumed that you have already successfully installed the Nominatim
6 software itself, if not return to the [installation page](Installation.md).
8 ## Configuration setup in settings/local.php
10 The Nominatim server can be customized via the file `settings/local.php`
11 in the build directory. Note that this is a PHP file, so it must always
16 without any leading spaces.
18 There are lots of configuration settings you can tweak. Have a look
19 at `settings/default.php` for a full list. Most should have a sensible default.
23 If you plan to import a large dataset (e.g. Europe, North America, planet),
24 you should also enable flatnode storage of node locations. With this
25 setting enabled, node coordinates are stored in a simple file instead
26 of the database. This will save you import time and disk storage.
27 Add to your `settings/local.php`:
29 @define('CONST_Osm2pgsql_Flatnode_File', '/path/to/flatnode.file');
31 Replace the second part with a suitable path on your system and make sure
32 the directory exists. There should be at least 64GB of free space.
34 ## Downloading additional data
36 ### Wikipedia/Wikidata rankings
38 Wikipedia can be used as an optional auxiliary data source to help indicate
39 the importance of OSM features. Nominatim will work without this information
40 but it will improve the quality of the results if this is installed.
41 This data is available as a binary download:
43 cd $NOMINATIM_SOURCE_DIR/data
44 wget https://www.nominatim.org/data/wikimedia-importance.sql.gz
46 The file is about 400MB and adds around 4GB to Nominatim database.
49 If you forgot to download the wikipedia rankings, you can also add
50 importances after the import. Download the files, then run
51 `./utils/setup.php --import-wikipedia-articles`
52 and `./utils/update.php --recompute-importance`.
54 ### Great Britain, USA postcodes
56 Nominatim can use postcodes from an external source to improve searches that
57 involve a GB or US postcode. This data can be optionally downloaded:
59 cd $NOMINATIM_SOURCE_DIR/data
60 wget https://www.nominatim.org/data/gb_postcode_data.sql.gz
61 wget https://www.nominatim.org/data/us_postcode_data.sql.gz
63 ## Choosing the Data to Import
65 In its default setup Nominatim is configured to import the full OSM data
66 set for the entire planet. Such a setup requires a powerful machine with
67 at least 64GB of RAM and around 800GB of SSD hard disks. Depending on your
68 use case there are various ways to reduce the amount of data imported. This
69 section discusses these methods. They can also be combined.
73 If you only need geocoding for a smaller region, then precomputed extracts
74 are a good way to reduce the database size and import time.
75 [Geofabrik](https://download.geofabrik.de) offers extracts for most countries.
76 They even have daily updates which can be used with the update process described
78 [other providers for extracts](https://wiki.openstreetmap.org/wiki/Planet.osm#Downloading).
80 Please be aware that some extracts are not cut exactly along the country
81 boundaries. As a result some parts of the boundary may be missing which means
82 that Nominatim cannot compute the areas for some administrative areas.
84 ### Dropping Data Required for Dynamic Updates
86 About half of the data in Nominatim's database is not really used for serving
87 the API. It is only there to allow the data to be updated from the latest
88 changes from OSM. For many uses these dynamic updates are not really required.
89 If you don't plan to apply updates, the dynamic part of the database can be
90 safely dropped using the following command:
93 ./utils/setup.php --drop
96 Note that you still need to provide for sufficient disk space for the initial
97 import. So this option is particularly interesting if you plan to transfer the
98 database or reuse the space later.
100 ### Reverse-only Imports
102 If you only want to use the Nominatim database for reverse lookups or
103 if you plan to use the installation only for exports to a
104 [photon](https://photon.komoot.de/) database, then you can set up a database
105 without search indexes. Add `--reverse-only` to your setup command above.
107 This saves about 5% of disk space.
109 ### Filtering Imported Data
111 Nominatim normally sets up a full search database containing administrative
112 boundaries, places, streets, addresses and POI data. There are also other
113 import styles available which only read selected data:
115 * **settings/import-admin.style**
116 Only import administrative boundaries and places.
117 * **settings/import-street.style**
118 Like the admin style but also adds streets.
119 * **settings/import-address.style**
120 Import all data necessary to compute addresses down to house number level.
121 * **settings/import-full.style**
122 Default style that also includes points of interest.
123 * **settings/import-extratags.style**
124 Like the full style but also adds most of the OSM tags into the extratags
127 The style can be changed with the configuration `CONST_Import_Style`.
129 To give you an idea of the impact of using the different styles, the table
130 below gives rough estimates of the final database size after import of a
131 2018 planet and after using the `--drop` option. It also shows the time
132 needed for the import on a machine with 64GB RAM, 4 CPUS and SSDs. Note that
133 the given sizes are just an estimate meant for comparison of style requirements.
134 Your planet import is likely to be larger as the OSM data grows with time.
136 style | Import time | DB size | after drop
137 ----------|--------------|------------|------------
138 admin | 5h | 190 GB | 20 GB
139 street | 42h | 400 GB | 180 GB
140 address | 59h | 500 GB | 260 GB
141 full | 80h | 575 GB | 300 GB
142 extratags | 80h | 585 GB | 310 GB
144 You can also customize the styles further. For a description of the
145 style format see [the development section](../develop/Import.md).
147 ## Initial import of the data
149 !!! danger "Important"
150 First try the import with a small extract, for example from
151 [Geofabrik](https://download.geofabrik.de).
153 Download the data to import and load the data with the following command
154 from the build directory:
157 ./utils/setup.php --osm-file <data file> --all 2>&1 | tee setup.log
160 ***Note for full planet imports:*** Even on a perfectly configured machine
161 the import of a full planet takes at least 2 days. Once you see messages
162 with `Rank .. ETA` appear, the indexing process has started. This part takes
163 the most time. There are 30 ranks to process. Rank 26 and 30 are the most complex.
164 They take each about a third of the total import time. If you have not reached
165 rank 26 after two days of import, it is worth revisiting your system
166 configuration as it may not be optimal for the import.
168 ### Notes on memory usage
170 In the first step of the import Nominatim uses osm2pgsql to load the OSM data
171 into the PostgreSQL database. This step is very demanding in terms of RAM usage.
172 osm2pgsql and PostgreSQL are running in parallel at this point. PostgreSQL
173 blocks at least the part of RAM that has been configured with the
174 `shared_buffers` parameter during [PostgreSQL tuning](Installation#postgresql-tuning)
175 and needs some memory on top of that. osm2pgsql needs at least 2GB of RAM for
176 its internal data structures, potentially more when it has to process very large
177 relations. In addition it needs to maintain a cache for node locations. The size
178 of this cache can be configured with the parameter `--osm2pgsql-cache`.
180 When importing with a flatnode file, it is best to disable the node cache
181 completely and leave the memory for the flatnode file. Nominatim will do this
182 by default, so you do not need to configure anything in this case.
184 For imports without a flatnode file, set `--osm2pgsql-cache` approximately to
185 the size of the OSM pbf file (in MB) you are importing. Make sure you leave
186 enough RAM for PostgreSQL and osm2pgsql as mentioned above. If the system starts
187 swapping or you are getting out-of-memory errors, reduce the cache size or
188 even consider using a flatnode file.
190 ### Verify import finished
192 Run this script to verify all required tables and indices got created successfully.
195 ./utils/check_import_finished.php
198 ### Setting up the website
200 Run the following command to set up the `website/settings-frontend.php`.
201 These settings are used in website/*.php files. You can use the website only after this
205 ./utils/setup.php --setup-website
208 This step is not necessary if you use `--all` option while setting up the DB.
210 ## Tuning the database
212 Accurate word frequency information for search terms helps PostgreSQL's query
213 planner to make the right decisions. Recomputing them can improve the performance
214 of forward geocoding in particular under high load. To recompute word counts run:
217 ./utils/update.php --recompute-word-counts
220 This will take a couple of hours for a full planet installation. You can
221 also defer that step to a later point in time when you realise that
222 performance becomes an issue. Just make sure that updates are stopped before
223 running this function.
225 If you want to be able to search for places by their type through
226 [special key phrases](https://wiki.openstreetmap.org/wiki/Nominatim/Special_Phrases)
227 you also need to enable these key phrases like this:
229 ./utils/specialphrases.php --wiki-import > specialphrases.sql
230 psql -d nominatim -f specialphrases.sql
232 Note that this command downloads the phrases from the wiki link above. You
233 need internet access for the step.
236 ## Installing Tiger housenumber data for the US
238 Nominatim is able to use the official [TIGER](https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html)
239 address set to complement the OSM house number data in the US. You can add
240 TIGER data to your own Nominatim instance by following these steps. The
241 entire US adds about 10GB to your database.
243 1. Get preprocessed TIGER 2019 data and unpack it into the
244 data directory in your Nominatim sources:
247 wget https://nominatim.org/data/tiger2019-nominatim-preprocessed.tar.gz
248 tar xf tiger2019-nominatim-preprocessed.tar.gz
250 `data-source/overview.md` explains how the data got preprocessed.
252 2. Import the data into your Nominatim database:
254 ./utils/setup.php --import-tiger-data
256 3. Enable use of the Tiger data in your `settings/local.php` by adding:
258 @define('CONST_Use_US_Tiger_Data', true);
260 4. Apply the new settings:
263 ./utils/setup.php --create-functions --enable-diff-updates --create-partition-functions