7.13. Analytic Querying¶
GeoMesa provides advanced query capabilities through GeoTools query hints. You can use these hints to control various aspects of query processing, or to trigger distributed analytic processing. See Setting Query Hints for details on setting query hints.
7.13.1. Feature Sampling¶
Instead of returning all features for a query, GeoMesa can use statistical sampling to return a certain percentage of results. This can be useful when rendering maps, or when there are too many features to be meaningful.
Features can either be sampled absolutely, or sampled by a certain attribute. For example, given a series of points in a track, you may wish to sample by the track identifier so that no tracks are completely sampled out.
The sampling value should be a float in the range (0, 1), which represents the fractional value of features that will be returned. Due to distributed processing, the actual count returned is not guaranteed to equal the requested percentage - however, there will never be less features than requested. For example, if you sample 5 features at 10%, you will get back anywhere from 1 to 5 features, depending on how your data is distributed in the cluster.
Sampling can also be combined with the other analytic queries mentioned below.
|QueryHints.SAMPLE_BY||String - attribute name (optional)||any string|
7.13.2. Density Queries¶
To populate heatmaps or other pre-rendered maps, GeoMesa can use server-side aggregation to map features to pixels. This results in much less network traffic, and subsequently much faster queries.
The result from a density query is an encoded iterator of
(x, y, count), where
y refer to
the coordinates for the center of a pixel. In GeoServer, you can use the WPS DensityProcess to create a
heatmap from the query result. See Heatmaps for more information.
7.13.3. Statistical Queries¶
GeoMesa supports generating various statistics against a data set. These stats are generated in a distributed scan, so provide built-in parallelism and require less network traffic. The following stats are supported:
- min/max values (bounds)
- enumeration of values
- top-k values
- frequency of values
- histogram of values
- descriptive statistics
In GeoServer you can use the
StatsProcess. Otherwise, the query is controlled through the
following query hints:
See Analytic Commands for information on running statistical queries through the GeoMesa command-line tools.
184.108.40.206. Explanation of Hints¶
This hint is a string describing the stats to be collected. Each type of stat has a corresponding string
representation. Multiple stats can be collected at once by delimiting them with a semi-colon. Instead
of constructing stat strings by hand, there are convenience methods in
that will generate valid stat strings. Stat strings can be validated by trying to parse them with
org.locationtech.geomesa.utils.stats.Stat.apply. The implementing classes are contained in the package
Stat strings are as follows:
|frequency (by time period)||Frequency||
As seen in the table above, multiple stats can be calculated at once through comma delimiting. In addition,
stats can be calculated on grouped values by using
GroupBy on a nested stat expression.
The Z3 frequency and histogram are special stats that will operate on the Z3 value created from the geometry and date.
<time period> can be one of
year, and indicates how data should be grouped.
<precision> for frequencies is defined as:
- for geometry and Z3 types, it is the number of bits of z-index to keep (max of 64). Note that the first 2 bits do not hold any information
- for date types, it is the number of milliseconds to group for binning
- for number types, it is the number of digits that will be grouped together
- for floating point types, it is the number of decimal places that will be considered
- for string types, it is the number of characters that will be considered
<bins> for a histogram indicate how many groupings should be made. The
set the initial sizes of the groupings, but are not hard limits. The histogram will expand if needed as
new values are added, but some precision may be lost.
This hint controls whether the stat will be returned as a serialized (encoded) object, or as a JSON string.
Serialized stats can be deserialized using an instance of
obtained through its factory
220.127.116.11. Accessing Stats through the GeoMesa API¶
In addition to queries through the GeoTools API, stats can be accessed directly through the GeoMesa API. Most
GeoMesa datastores implement
org.locationtech.geomesa.index.stats.HasGeoMesaStats, which defines a single method:
def stats: org.locationtech.geomesa.index.stats.GeoMesaStats
In addition to running queries, the
GeoMesaStats interface can be used to retrieve cached stats.
See Configuring Cached Statistics for details on configuring cached stats.
7.13.4. Arrow Encoding¶
GeoMesa supports returning features as Apache Arrow encoded vectors. This provides an optimized columnar memory layout for fast processing and interoperability with other systems.
The result of an Arrow query will be an iterator of SimpleFeatures, where the first attribute of each will be a byte array. Concatenated together, the byte arrays will form an Arrow file, in the Arrow streaming format (i.e. no footer).
In GeoServer you can use the
ArrowConversionProcess, or through WFS by setting
outputFormat=application/vnd.arrow and controlling the configuration through the
format_options=includeFids:true;batchSize:1000. Otherwise, the encoding is controlled through the
following query hints:
|Key||Type||GeoServer Format Option|
18.104.22.168. Explanation of Hints¶
This hint is used to trigger an Arrow query.
This hint controls whether to include the feature ID as an Arrow vector or not. The default is to include it.
This hint controls whether to return the full feature ID, or a 4-byte proxy ID. Proxy IDs can be used
for callbacks by using the
proxyID() CQL filter function.
This hint allows for sorting the results by a particular attribute. Only attribute names are supported, not arbitrary CQL.
This hint is used to flip sort order from normal (ascending) to reverse (descending).
This hint indicates which simple feature attributes should be dictionary encoded. It should be a comma-separated list of attribute names.
This hint indicates known dictionary values to use for encoding each field. This allows for specifying a known dictionary up front, which means the dictionary doesn’t have to be computed. Values which are not indicated in the dictionary will be grouped under ‘other’.
The hint should be an encoded map of attribute names to attribute values. The hint should be encoded in
comma-separated values format, where each line indicates a different attribute. The first item in each line is
the attribute name, and the subsequent items are dictionary values. Standard CSV escaping can be used. The function
org.locationtech.geomesa.utils.text.StringSerialization.encodeSeqMap can be used to encode a map of values.
This hint indicates that cached statistics (top-k) will be used for dictionaries, if available. Otherwise, dictionaries will be computed based on the data returned, which may be slower.
This hint will cause multiple logical Arrow files to be returned, instead of a single file. This will generally be faster, as no client-side merging needs to be done. However, any sorting will only be applied per file, not globally. Also, the end result tends to be larger (in bytes), as metadata and dictionary values may be repeated in different logical files.
This hint will cause any dictionaries to be computed first, through a separate scan. A second scan will construct the Arrow files. This is the behavior of the initial GeoMesa Arrow implementation, and is only included for back compatibility.
This hint will restrict the number of features included in each Arrow record batch. An Arrow file contains a series of record batches - limiting the max size of each batch can allow memory-constrained systems to operate more easily.
This hint controls the IPC format version for Arrow binary encoding. It should be a valid Arrow format version,
0.10. The Arrow IPC format changed slightly starting with version
22.214.171.124. Example Query¶
7.13.5. Binary Encoding¶
GeoMesa supports returning features in a custom binary format (referred to as BIN) that uses 16 or 24 bytes per feature. This provides an extremely compact representation of a few key attributes.
The 16 byte BIN format is as follows:
<4 byte int><4 byte int><4 byte floating point><4 byte floating point>
The first integer is referred to as a track ID, and is generally used to group related points. For example, a line string may be turned into several BIN records with a common track ID. The second integer is a date represented as the number of seconds since the Java epoch (Jan. 1, 1970). The two floating point numbers represent the latitude and longitude of the record, respectively.
The 24 byte BIN format is the same as the 16 byte version, but with an additional 8 bytes at the end for arbitrary data.
The result of a BIN query will be an iterator of SimpleFeatures, where the first attribute of each will be a byte array containing one or more BIN-encoded features.
In GeoServer you can use the
BinConversionProcess. Otherwise, the encoding is controlled through the
following query hints:
126.96.36.199. Explanation of Hints¶
This hint is used to trigger a BIN query. It should be the name of an attribute that will be used to generate the track ID for each record.
This hint controls the geometry attribute used for each record. If omitted, the default geometry of the feature type is used.
This hint controls the date attribute used for each record. If omitted, the default date of the feature type is used.
This hint will trigger the creation of 24-byte records, instead of the standard 16. It should be the name of an attribute that will be used to generate the label for each record.
This hint will cause the records to be sorted. It should be the name of an attribute in the feature type.
This hint controls the batch size used when generating BIN records.