Fri, 2016-01-22 17:59

A Platform for Data Interoperability: Koop

There is no shortage of geographic data available on the web with more source coming online all the time. The downside of this explosion is that new sources often have their own API and proprietary formats that are not compatible with each other. A single data source on it's own is never as useful as when it is viewed in context.

What is needed is a platform for interoperability. Koop steps into this challenge by treating GeoJSON as a kind of compile target. As long as data can be fetched and translated, any dataset from any source can be viewed alongside any other.

Fri, 2016-01-22 15:28

Using the FGDC Address Standard with Non-English Street Names

Street names carry local context and meaning and can vary locally, even within a relatively small geographic envelope, frequently appearing in languages other than English. The United States Federal Geographic Data Committee (FGDC) in 2011 released a data content standard for storing address data in a highly normalized way. This standard was intended for use within the United States of America and principally for parsing English language street names. This session will look at how to use this the FGDC address data standard for processing and storing non-English street names.

Thu, 2016-01-21 17:02

Don't Forget the Elephant

Many database administrators were first introduced to PostgreSQL years ago and have memories (good & bad) of using version numbers starting with 6, 7 & 8. Many people are still running these old versions as well, nervous about upgrading and unsure of the benefits. With the rise in popularity of MySQL, many open source advocates turned there since the majority of internet community support was centered around it. And the advanced features that commercial databases such as Oracle & SQL Server were advertising appealed to many enterprise administrators.

Wed, 2016-01-20 04:50

BRIN indexes on geospatial database

BRINs (Block Range INdexes) are one of the major new features introduced with PostgreSQL 9.5. They allow to directly select just blocks of table pages needed for queries execution, resulting on smaller indexes that can be easily contained in memory and that require less maintenance than the existing ones.

These features make BRIN particularly suitable for very large tables, and more in general for query with a low-selectivity that would be executed preferring a sequential scan of all tables blocks instead of using indexes.

Tue, 2016-01-19 14:25

Machine Learning on Geospatial Datasets for Segmentation, Prediction and Modeling

Machine learning is powerful technique that allows us to create predictive data driven models that can learn off complex multivariable data. The geospatial world is full of such datasets where its hard to know exactly how the input variables to your model will effect the outcomes. There exists a growing ecosystem of libraries and frameworks like Tensor Flow and Scikit-Learn that allow for sophisticated machine learning to take place but very few are easily interoperable with geospatial frameworks like PostgreSQL..

Tue, 2016-01-19 13:44

Wrapping up Python into a Cloud-based PostgreSQL

Through PostgreSQL’s support for user-defined procedural language functions, the possibilities for data analysis within a database is greatly expanded. Specifically, using PL/Python, one can bring in countless Python libraries to process data close to the database. Here I will talk about my efforts to bring in the functionality of PySAL, a spatial analytics library written in Python and developed largely by Serge Rey, et al. at Arizona State University.

Tue, 2015-11-24 01:40

Running autoscaling Postgresql at Scale!

You keep hearing about containers and maybe you have even played with Docker. You are now wondering - how do I run and manage this in production? In this session we are going to show you how. We'll level set with a quick intro to Docker, then show how CrunchyDB has taken advantage of containers to do more with Postgresql. Then we will demo bringing these images up to scale and orchestrated with Kubernetes and OpenShift - two opensource project used to manage containers.

Fri, 2015-11-13 15:33

Workshop: Building Dynamic Maps with CartoDB

In this 4 hour intensive hands-on workshop, we will learn how to use raw data to create beautiful, interactive, and sharable maps for the web. While this session does not require any technical skills, participants will quickly learn the fundamentals of making maps in the CartoDB editor, and progress into advanced toics such as using Torque.js to create animated visualizations, data-driven design in CartoCSS, spatial queries and spatial analysis using PostgreSQL & PostGIS, designing interactivity using HTML, and basic principles of data visualization and cartographic theory.

Subscribe to PostgreSQL