14.3. Using the HBase Data Store Programmatically¶
14.3.1. Creating a Data Store¶
An instance of an HBase data store can be obtained through the normal GeoTools discovery methods, assuming that the GeoMesa code is on the classpath.
The HBase data store also requires that an appropriate hbase-site.xml
file is available on the classpath; the
connection parameters for HBase are obtained from this file, including hbase.zookeeper.quorum
and
hbase.zookeeper.property.clientPort
.
As an alternative to providing hbase-site.xml
, the Zookeeper connection can be specified through the
parameter hbase.zookeepers
. However, this method is not recommended, as other important configurations
(including security, if any) from hbase-site.xml
may be required for correct operation.
Map<String, Serializable> parameters = new HashMap<>();
parameters.put("hbase.catalog", "geomesa");
org.geotools.data.DataStore dataStore =
org.geotools.data.DataStoreFinder.getDataStore(parameters);
14.3.2. HBase Data Store Parameters¶
The data store takes several parameters (required parameters are marked with *
):
Parameter |
Type |
Description |
---|---|---|
|
String |
The name of the GeoMesa catalog table, including the HBase namespace (if any) separated
by a |
|
String |
A comma-separated list of servers in the HBase zookeeper ensemble. This is optional,
the preferred method for defining the HBase connection is with |
|
String |
Path to the GeoMesa jar containing coprocessors, for auto registration |
|
String |
Additional HBase configuration resource files (comma-delimited) |
|
String |
Additional HBase configuration properties, as a standard XML |
|
Boolean |
Re-use and share HBase connections, or create a new one for this data store |
|
Boolean |
Can be used to disable remote filtering and coprocessors, for environments where custom code can’t be installed |
|
Boolean |
Enable HBase security (visibilities) |
|
String |
Comma-delimited superset of authorizations that will be used for queries |
|
Boolean |
Forces authorizations to be empty |
|
Boolean |
Audit queries being run. Queries will be written to a log file |
|
String |
The max time a query will be allowed to run before being killed. The
timeout is specified as a duration, e.g. |
|
Integer |
The number of threads to use per query |
|
Integer |
The number of HBase RPC threads to use per coprocessor query |
|
Boolean |
Use loose bounding boxes - queries will be faster but may return extraneous results |
|
Integer |
Max ranges per extended scan. Ranges will be grouped into scans based on this setting |
|
Integer |
Max ranges per coprocessor scan. Ranges will be grouped into scans based on this setting |
|
Boolean |
Disable coprocessor scans for Arrow queries, and use local encoding instead |
|
Boolean |
Disable coprocessor scans for Bin queries, and use local encoding instead |
|
Boolean |
Disable coprocessor scans for density queries, and use local processing instead |
|
Boolean |
Disable coprocessor scans for stat queries, and use local processing instead |
|
Boolean |
Toggle coprocessors yielding partial results |
|
Boolean |
Toggle extremely parallel coprocessor scans (bounded by RPC threads) |
|
Boolean |
Toggle collection of statistics (currently not implemented) |
|
Boolean |
For partitioned schemas, execute scans in parallel instead of sequentially |
|
Boolean |
Toggle caching of results |
Note: the hbase.coprocessor.*.enable
parameters will be superseded by hbase.remote.filtering=false
.
More information on using GeoTools can be found in the GeoTools user guide.