On Geospatial Data and Solutions. Q&A with Glenn Stowe
” The biggest challenges usually come from data volume. Geospatial datasets can be massive – terabytes to petabytes. You need efficient and scalable storage and processing mechanisms in order to take full advantage of them. “
Q1. What are the typical applications that leverage geospatial data?
Common applications for mapping, navigation, ridesharing and delivery use geospatial data. There’s also an earth observation category of application that includes resource monitoring, weather forecasting, climate change mitigation, defense and intelligence. Insurance companies use applications with geospatial data for fire and flood risk management, and the construction industry has a whole range of applications, from job site monitoring to surveying to cost estimation. Business intelligence and commercial location planning uses are popular too. The range of use cases that unlock business and social value is potentially enormous.
Q2. What are the key challenges for businesses to integrate geospatial data into their applications to offer services and capabilities that they couldn’t previously?
Geospatial data comes in an array of formats, many of them dictated by historical tech trends like the dominant desktop platforms of the past. A long history of closed ecosystems and proprietary software made interoperability a challenge. As software development shifted from desktop to web applications, which are distributed by nature, this became a critical flaw. However, the biggest challenges usually come from data volume. Geospatial datasets can be massive – terabytes to petabytes. You need efficient and scalable storage and processing mechanisms in order to take full advantage of them.
Q3. If we look at the various possible use cases — from mapping services for everything from location awareness to route optimization, to applications that improve forecast accuracy, inventory management, and even equity trading — what are the common challenges posed by such applications?
From a platform perspective, the biggest challenges usually revolve around getting to the data quickly and being able to make some sense of it in a timeframe that is acceptable to the user. Research tells us that people expect a response from an action in a web application within 3 seconds. Any longer than that and the user’s perception of the app suffers. When you need to crunch hundreds of satellite images or scan through millions of streets in a database, this can be difficult.
Q4. MariaDB® Corporation has recently acquired CubeWerx, leaders in geospatial solutions. What kind of cloud-native, scalable geospatial capabilities are you planning to add to your database products?
We are focused on integrating CubeWerx technology into SkySQL, MariaDB’s cloud services platform to be able to manage and leverage very large geospatial datasets through standards-based REST APIs. These APIs are already supported by the tools programmers use. With this approach, MariaDB will have the capability to deal with large satellite and aerial image collections, and will leverage native loaders to handle most existing geospatial data formats.
Q5. You are planning to use a cloud-native approach. Why not add geospatial capabilities directly into the database, like other database vendors do?
If the cloud had existed when other vendors were deciding on their geospatial strategy, they probably would not have added those capabilities directly into the database. The cloud has really changed everything, including how we think about data management. The newer geospatial data formats are truly cloud native. They’re designed to be accessed directly from cloud object storage, including random access, using HTTP range requests. This adds a lot of possibilities for scalability, microservices, etc., that would not be possible through a traditional database pipeline.
Q6. How does MariaDB’s vision for how to deliver geospatial capabilities align with what they have built at CubeWerx?
CubeWerx team members were the original developers behind Oracle Spatial. The evolution of CubeWerx technology started with two questions: What if developers didn’t have to do everything in the database? What would they do differently? The direction was to build a platform that used the best of what the database could provide with the best of what could be done in an application server. That eventually evolved into applying the best of what the cloud could bring to the equation. MariaDB sees that this is the future of geospatial application development, making it extremely easy to leverage rich geospatial data through open standard APIs. MariaDB’s highly-scalable cloud platform SkySQL is the catalyst for letting all this come together.
Q7. How does it differ from Oracle Spatial?
That’s sort of comparing apples to oranges. Oracle Spatial is a geospatial extension to the Oracle database. There are tools they’ve added on top of that, like Java-based web services, etc., but it doesn’t form an integrated platform for building modern, REST-based web applications. With the acquisition of CubeWerx, MariaDB is taking a web-first approach. The entire architecture is based around providing geospatial capabilities through web services.
Q8. Tell us about the implication for geospatial to use an open standard API-driven approach?
The goal here is to provide geospatial the way developers use geospatial. The approach to building geo-enabled applications has shifted over the years, from a database-centric model to a web services one. The front end app developers don’t need to know or care how the data is stored and organized. They simply want the REST endpoints. MariaDB’s approach gives them that. Using open-standards-based APIs means that every existing desktop tool and web mapping toolkit already speaks to them, and nearly every geospatial developer is already familiar with them.
Q9. MariaDB® Corporation announced it has joined the Open Geospatial Consortium (OGC). What is it? and what do you expect from this membership?
The OGC is the standards body for all aspects of geospatial, but particularly for APIs and cloud representations of geospatial data. As a big proponent of open source software, supporting open API standards is a natural move for MariaDB, particularly in this case. The CubeWerx team have been leaders in the development of those standards for over 25 years and their platform uses those APIs exclusively. MariaDB is expanding the CubeWerx membership in order to remain actively involved in OGC’s evolution.
Qx Anything else you wish to add?
MariaDB’s developer-centric approach to delivering geospatial capabilities is one that no other database vendor offers, but it’s the approach that opens up the greatest possibilities and efficiencies for building applications. With this modern cloud-native approach, companies can manage virtually infinite amounts of geospatial data on low-cost, durable cloud storage and provide OGC standards-based REST APIs to access the data.
Glenn Stowe has over 30 years of experience in the geospatial industry with a strong focus on standards-based web services and scalable solutions to big-data problems. He is the product manager for geospatial at MariaDB Corporation and before that was co-founder and vice president at CubeWerx, an industry leader in developing standards and software for open geospatial web services. He lives and works in Val-Des-Monts, Canada with his wife Pamela and their rottweiler, Rocky.
Sponsored by MariaDB Corporation