Spatial: What’s New in Oracle Database 19c?

Copyright © 2019, Oracle and/or its affiliates. All rights reserved.
Oracle Ask TOM Office Hours:
Spatial: What’s New in
Oracle Database 19c?
Siva Ravada, Senior Director of Development, Oracle Spatial and Graph
Jean Ihm, Product Manager @JeanIhm

Copyright © 2019, Oracle and/or its affiliates. All rights reserved. |
Safe Harbor Statement
The following is intended to outline our general product direction. It is intended for
information purposes only, and may not be incorporated into any contract. It is not a
commitment to deliver any material, code, or functionality, and should not be relied upon
in making purchasing decisions. The development, release, timing, and pricing of any
features or functionality described for Oracle’s products may change and remains at the
sole discretion of Oracle Corporation.
2

AskTOM sessions on Spatial features
• Welcome to our 1st AskTOM Spatial session
• Visit the Spatial and Graph landing page to view recordings of past
sessions; submit feedback, questions, topic requests; view upcoming
session dates and topics; sign up
3
https://devgym.oracle.com/pls/apex/dg/office_hours/3084

Strategic Vision: The Platform
• Ubiquitous Spatial services in IT infrastructure
– Dramatically lower costs
– Simplify application development
– Integrate operational systems
• Allow Spatially-enabled solutions to focus on business context, not infrastructure
services
– Database and Application Server manage deployment infrastructure
• Security • Scalability • Standards Compliance
• Load Balancing • Failover • HW/SW Dependencies
– Support multiple application models with common data model
• Java • GML • Future …
• .Net • Web Services/SOA

Oracle Spatial and Graph
Location and graph analysis with secure storage for enterprise data
Topologies
3D / LiDAR
Raster
Networks
Web Services (OGC)Geocoding RoutingDeployable Services
RDF Graphs Property Graphs
Mapping
Polygons
Lines
Points
Location Tracking
(Geofencing)
Studio

Oracle Spatial and Graph
On Premises, Cloud and in Autonomous Database
6

Oracle’s Spatial Analytics Platforms
Oracle Big Data
Spatial and Graph
Oracle Database
Spatial and Graph

Ease of Use
• JSON and Oracle REST
Data Services
improvements
• Improved web services
user interface, CSW and
WFS enhancements
• Self-service development
tool
Performance
• Ability to use spatial
operators without a
spatial index
• Spatial index performance
improvements
– Enhancements to CBTree
index to use the data layer
directly for Spatial index
access.
– 3x faster query performance
for large point data sets.
8
Improved Database
Support
• Spatial support for
distributed transactions
• Spatial support for
database sharding
• Spatial Studio
Major New Spatial Features in 19c

GeoJSON support (introduced in 12.2)
• select json_value('{"type": "Point",
"coordinates": [125.6, 10.1]}', '$'
returning sdo_geometry) from dual
• Extend JSON support in the
database with Spatial operations
• JSON_VALUE() to support
GeoJSON and SDO_GEOMETRY
• SDO_GEOMETRY constructors
extended to take JSON as input
• Support spatial index and spatial
queries on JSON documents

Spatial JSON support
• New JSON support in addition to GeoJSON
– The GeoJSON standard does not allow all the types and coordinate systems we support with SDO_GEOMETRY
– New JSON support is introduced to support all the data we can represent with SDO_GEOMETRY (both from and to
functions); this is in addition to the GeoJSON support introduced in 12.2.0.1
2D polygon: {"polygon": {"boundary": [{"line": {"datapoints": [[3.0, 0.0], [6.0, 0.0], [9.0, 3.0], [9.0, 6.0], [6.0,
9.0], [3.0, 9.0], [0.0, 6.0], [0.0, 3.0], [3.0, 0.0]]}}, {"line": {"datapoints": [[4.0, 2.0], [3.0, 3.0], [3.0, 4.0], [4.0,
5.0], [5.0, 5.0], [6.0, 4.0], [6.0, 3.0], [5.0, 2.0], [4.0, 2.0]]}}]}}
Voided Polygon (LRS 3303): Geodedic i/o rings connected by lines{"srid": 8307, "spatialdimension": 2,
"polygon": {"boundary": [{"line": {"datapoints": [[3.0, 0.0, 0.0], [6.0, 0.0, 3.0], [9.0, 3.0, 7.24264069], [9.0,
6.0, 10.2426407], [6.0, 9.0, 14.4852814], [3.0, 9.0, 17.4852814], [0.0, 6.0, 21.7279221], [0.0, 3.0,
24.7279221], [3.0, 0.0, 28.9705627]]}}, {"line": {"datapoints": [[4.0, 2.0, 28.9705627], [3.0, 3.0, 30.3847763],
[3.0, 4.0, 31.3847763], [4.0, 5.0, 32.7989899], [5.0, 5.0, 33.7989899]]}}]}}
• ORDS Support for Spatial operations is enabled using this feature

Spatial operators without the index
• Data stored in an external table using text format for geometry
– NFS file, HDFS, Object Store, etc.
• Data from a JOIN view or some other complex view
• Spatial index not mandatory for using spatial operators
– Queries that use SDO_FILTER/SDO_RELATE, etc. can now be used even when there is no index on the table
– Spatial index is recommended when querying against the tables, but when queries need to be written against
views where the geometry data is constructed on the fly, this enhancement allows the queries to be written the
same way as if the data is coming from a table

Spatial operators without the index
• Prior to 18c
• In 18c
• Turn this off with a DB init parameter if old behavior is required
Oracle Confidential – Internal/Restricted/Highly Restricted
select sdo_geom.SDO_DISTANCE(ci.geometry, SDO_GEOMETRY(tw.geometry, 4326), 0.05, 'UNIT=MILE'),
ci.name, tw.user_id
from CINEMA ci, TWEETS_EXT_TAB_FILE tw
where sdo_geom.WITHIN_DISTANCE(ci.geometry, 0.25, SDO_GEOMETRY(tw.geometry, 4326), 0.05,
'UNIT=MILE') = 'TRUE'
select sdo_geom.SDO_DISTANCE(ci.geometry, SDO_GEOMETRY(tw.geometry, 4326), 0.05, 'UNIT=MILE'),
ci.name, tw.user_id
from CINEMA ci, TWEETS_EXT_TAB_FILE tw
where SDO_WITHIN_DISTANCE(ci.geometry, SDO_GEOMETRY(tw.geometry, 4326),
'DISTANCE=0.25 UNIT=MILE') = 'TRUE'

New Index Type Optimized for Point Data
• R-tree index has performance bottlenecks when large numbers of
concurrent DMLs are done on the spatial tables
• Spatial index based on B-tree index
– Internally create a composite b-tree index on (x,y) or (x,y,z)
• Improves DML performance for large volumes of updates for point only
data
• Queries are still done using the same standard syntax, so no application
changes are required
13
CREATE INDEX pt_idx on PT_CB(c2) indextype is mdsys.spatial_index_v2
PAREMETERS ('layer_gtype=POINT cbtree_index=true');

CB-tree index for point data
• CB-Tree enhancements in 19c
– Enhanced with internal data layer interfaces
– 3X performance improvement for queries
– Supports distributed transactions
– Orders of magnitude faster for high volume DML operations
select count(*) from abi, states where sdo_filter(geometry, geom) = 'TRUE';
12.2.0.1: Elapsed: 00:00:14.66
18.1.0.0: Elapsed: 00:00:04.70

• The use of R-tree spatial indexes is not
supported in distributed and Oracle XA
transactions
– This is a restriction prior to the 18.1 release
• Enables web applications that modify spatial
data
• What we did
– Use regular tables instead of global temporary tables
for managing the transaction data when the
transaction is marked as XA
Support distributed transactions

• Oracle Sharding is a scalability and
availability feature for custom-
designed OLTP applications that
enables distribution and replication
of data across a pool of Oracle
databases that do not share
hardware or software
• The pool of databases is presented
to the application as a single logical
database.
16
Sharding in the DB

Spatial Specific Sharding concepts
• USER_SDO_GEOM_METADATA is created on each shard separately
– This is a pain now, we are looking to remove this restriction so that coordinator can do this
• All DDL (create index, alter index, etc.) are performed on the coordinator and they are
automatically propagated to the shards
• All queries that need to "cross-shards" are performed on the coordinator and are
automatically aggregated and shard-specific queries are performed on the individual
shards

Location tracking Server
• Track objects against a set of
polygons and generate
notifications when the moving
objects enter the regions of
interest
18
Introduced in 18c
• Track objects within a set of
polygons and generate
notifications when the moving
objects leaves the regions of
interest

Tracking Server
• Tracker object notification types
– ‘I’ alert when object enters a region
– ‘O’ alert when the object leaves a region
– ‘U’ alert whenever the object state changes from inside->outside or outside->inside
• Notification Message
– ‘E’ generates a message when the alert condition is true (continuous alert)
– ‘T’ generates a message when the alert condition flips between ‘I’ and ‘O’
• Tracking can be dynamically turned off for moving objects
– isActive: 'Y’ or ‘N’
• New Java API to enable Java applications
Oracle Confidential – Internal/Restricted/Highly Restricted 19

Location tracker Performance
• Linux Server: 8GB, 8 CPUs
• 9 regions, 900 objects tracked
– 100 steps in all regions, 810,000 polygon checks: (secs)40.42
– 1000 steps in all regions, 8,100,000 polygon checks: (min:sec) 6:08
– 10,000 steps in all regions, 81 million polygon checks: (min:sec) 48:34
– 100,000 steps all regions, 810 million polygon checks (hrs:min) 6:47
• A little math shows each object having its location updated around 4 times
a second
20

Oracle Map Visualization
• HTML5 based
• Deployed in J2EE server
• Enabling developers to incorporate
interactive maps and spatial analysis into
applications
• Integrating data from Oracle Spatial and
Graph or other sources (WMS, WFS,
GeoRSS, WMTS), and background maps
from data providers or services
• JavaScript, Java, and XML APIs for web
mapping apps

Map Visualization
• Dynamic Tile Layer
– Define tile layer dynamically in the V2 API (instead of pre-built in the server and
stored permanently in metadata view)
– Specify or change rendering style as needed
– Server generates tile images just like regular tile layer
• Supports large and complex queries, without unnecessary storage and
management overhead

Offline map cache
• Export and use tile images for offline usage
• User can select areas and zoom levels to generate the tiles
• The tiles can be compressed and downloaded
• V2 API changes have been made so that the tiles can be used even in off-
line mode (w/o connection to backend MapViewer server or database)

• Single layer vector tile generation
– MapViewer can now generate vector tiles in
the 'industry standard' vector tiles
– Requests look like this:
http://localhost:8080/mapviewer/vt/THEM
E_CUSTOMERS/8/41/99.mvt
– Any client mapping API that supports vector
tiles can utilize it (OpenLayers, MapBox GL JS)
– The screenshot showing the tax drop locations
(point dataset) as vector tiles
• Not supported for base maps
24
Vector Tiles

GeoRaster
• A data type to store raster data
– Satellite images, remote sensing data
– Multi-band, multi-layer
– An XML schema to store Metadata
• Data source, layer information
– Geo Referencing information
• Functionality
– storage and indexing of raster data
• No size limit for each raster object
– Generate resolution pyramid
– query and analysis
– delivering to external consumers
• Publish as JPEG, GIFF images

GeoRaster 18c New Features
• GDAL is Integrated into Oracle Database
– The new SDO_GEOR_GDAL Package integrates GDAL into Oracle database server
– It provides server-side raster data loading, exporting, and in-database terrain analysis capabilities
– It also enables and simplifies development of C/C++ plug-ins through the GDAL API
• Supports On-the-fly Image Processing
– equalize, normalize, piecewise stretch, stretch, and filter
• New Statistics on Virtual Mosaics
– New SDO_GEOR_AGGR.getMosaicStatistics Procedure generates statistics and histogram for virtual mosaics
enabling better visualization of virtual mosaics
• 4GB Limit Removed for External JPEG 2000 Image Files
– Large JPEG 2000 images can now be fast loaded into database without decompression
• Java API and GeoRaster Viewer Enhancements
– Java API supports all new features and the Viewer displays large images and virtual mosaics better

Example – Image Mosaic with Histogram Matching
27

GeoRaster 19c New Features
• Raster Algebra supports analytics at the pixel level
– Calculating the vegetation cover from a satellite image, calculating land use polygons from images, etc.
• Supports On-the-fly Raster Algebra
– All subprograms in the SDP_GEOR_RA package support storing the result data in temporary BLOB so that users
can do raster analysis and cartographic modeling on-the-fly to support various real time applications
– Supported in both PL/SQL API and JAVA API
• GeoRaster Must Be Enabled at Schema Level
– Previously GeoRaster is enabled for the whole database
– This enhances database security for the cloud

• OGC Standard web services: WFS, CSW, WCS, WMS, WTS, etc.
• Web catalogue Service 2.0.0.
– Based on new XML binary storage
– Supports DCMI, Inspire, ISO profiles
• Web Coverage Service 2.0.0., 2.0.1.
– Publish data stored in Georaster as coverages
• Improvements for WFS
• Unified console for web services
Web Services

Web Services Support
• Console Enhancements
– The Spatial Web Services administration console for WFS, WCS, and CSW is enhanced
with standard user interfaces, as well as multiple data sources management
capabilities
– Make it easy to publish vector and raster data using console menu options (no code
required)
• The OpenGIS Catalogue Services Specification (CSW) 2.0.2
– ISO Metadata Application Profile is supported
• Multiple data source support
– WFS/CS-W/WCS can support reading the data from multiple data sources in the same
WLS instance

OGC Services: Multiple data sources
• Multiple data source support
– WFS/CS-W/WCS can support reading the data from multiple data sources in the same
WLS instance
http://<machine-name:port>/oraclespatial/wfs/<data source
name>?request=GetCapabilities&service=WFS&version=1.1.0
31

Oracle Spatial and 3D
Solids
3D Queries
Triangular
Irregular Networks
Point Clouds
Import
and
Export
Spatial and
attribute queries
Visibility queriesVolumetric Analysis
3D Coordinate Systems

City Models – CityGML
• Supports
– CityGML 2.0
– 3dCityDB 4.0.1
– Core model, geometry representation, appearance model, buildings, city furniture
model, land use, transportation, vegetation, water body
• Validation
– Geometric
• Polygons (SURFACE_GEOMETRY), root geometries
– Semantic – not yet
• Targeting
– Large data sets, scalability
– Data integration
– Analytical options

Point Cloud on Non Exadata systems
• Hybrid Storage Model
– Flat table storage for easy load and update
– Clustered data for faster query
• Clustering helps group the nearby points in the same data blocks to reduce the I/O cost
– We can load LAS/LAZ files directly into this hybrid data model
– Queries do not need a spatial index so as soon as data is loaded we can make it
available for queries
• B-tree indexes are used for queries since it is only point data

Other Enhancements for Point Cloud data
• Pyramiding support for PC and TIN data
– PL/SQL API added to generate pyramids for existing point clouds
– Points can be repeated in the pyramid levels or mutually exclusive
– Useful for visualization applications
• Contour generation from PC data
– Generate linear geometries that connect points with equal elevation values
– A polygon can be specified to limit contour generation to a region
– Contour generation process is grid-based in x and y
– Krigging performed to assign heights to grids with no point data

Point-Cloud to Contour

PointCloud To DEM
Input: PointCloud
Output: GeoRaster

Copyright © 2019, Oracle and/or its affiliates. All rights reserved. | 38
One Autonomous Database - Optimized by Workload
Autonomous
Transaction Processing (ATP)
Best for all Analytic Workloads:
• Data Warehouse, Data Mart
• Data Lake, Machine Learning
Autonomous
Data Warehouse (ADW)
Best for TP and Mixed Workloads:
• Transactions, Batch, Reporting, IoT
• Application Dev, Machine Learning
ORACLE
AUTONOMOUS
DATABASE

ATP and Autonomous related changes
• In ATP, Data Vault is enabled by default to protect customer data
• This means, Oracle created accounts cannot read the customer data
• MDSYS can no longer directly manage the index, statistics, and other
metadata data stored in the user schemas
– Security model and package execution model has to be changed so that everything
gets executed as the user
• All the J2EE components have to support the secure TNS protocol by
default
• The J2EE components will be re-packaged as Cloud Market place
components
39

Spatial in ADW/ATP-Shared
• About 80% Spatial functionality in the DB is enabled by default
• No JAVAVM support in the current release
– The following are excluded due to no JAVAVM
• Topology
• Geocoder in the DB
• 3D functions (including LIDAR)
• Network Data Model and Routing Engine
• Following will be supported in 2020
• Spatial Visualization (MapViewer)
• OGC Services
• GeoRaster
• J2EE Geocoder
Confidential

Spatial in ATP-Dedicated
• All the Spatial features in the DB are supported
• All of the J2EE components will be shipped as plugins in Oracle Cloud
Market Place
– Work in progress to make these components deployable from Cloud Market place
• Following will be supported in 2020
• Spatial Visualization (MapViewer)
• OGC Services
• J2EE Geocoder
• Network Data Model
• Routing Engine
Confidential

Oracle Spatial and Graph:
Natively manage all spatial content in the DB
Topologies
f1
f2n1
n2
e1
e2
e3
e4
3D / LIDAR
Raster
Networks
Polygons
Lines
Points
Web Services (OGC)RoutingDeployable Services Mapping GeocodingOracle Cloud
Market Place
Autonomous DB

Introducing Spatial Studio
Self-service spatial analytics
https://www.oracle.com/database/technologies/spatial-studio.html

• 100’s of SQL
spatial analysis
operators
– Filter
– Combine
– Transform
– Measure
Confidential – Oracle Internal/Restricted/Highly Restricted
44
No code environment for Spatial Analysis

45
Resources

Resources

Resources
Product homepage: oracle.com/database/technologies/spatialandgraph.html
– Blog: blogs.oracle.com/oraclespatial
– Forum:
community.oracle.com/community/database/oracle-database-options/spatial
– Oracle Spatial and Graph User Group: linkedin.com/groups/1848520/
– YouTube Channel: youtube.com/c/OracleSpatialandGraph
– Twitter: @SpatialHannes @JeanIhm
Oracle Spatial and Graph – Spatial Features

AskTOM sessions on Spatial Features
• Next Spatial and Graph session in early August
–Topic: Oracle Spatial Studio
• View recordings, submit feedback, questions,
topic requests, view upcoming session dates and
topics, sign up to get regular updates
48

49
Thanks for attending! See you next time.

Spatial: What’s New in Oracle Database 19c?

Recomendados

Recomendados

Mais conteúdo relacionado

Último

Último (20)

Destaque

Destaque (20)

Spatial: What’s New in Oracle Database 19c?