Faster rendering for real-time, big data insights
By: John Goolgasian
From defense and intelligence agencies to travel, businesses, and financial services, Geospark Analytics helps organizations make sense of vast amounts of location-based data. Location context is critical for users to derive insights from Geospark Analytics’s extensive, real-time data sources, which is why the company’s risk and threat intelligence solution, Hyperion, just upgraded to our maps and geocoder. John Goolgasian, COO, tells us about their switch from Leaflet.
Recently, we changed the name of our analytics-as-a-service platform from BlueGlass to Hyperion. That transition was more than a name change; it signified the evolution of the service from a research program to an enterprise solution for risk and threat intelligence. Our clients will benefit from this transition with next-generation features, including a more dynamic and powerful mapping engine — enter Mapbox GL.
We were pushing, if not exceeding, the upper limits of Leaflet’s capabilities, the open-source interactive map library that we adopted when BlueGlass was an R&D effort. As a mature application, Hyperion needed enhanced capabilities to achieve the performance and flexibility our clients expect. We needed to make a transition, and the Mapbox platform was the easy choice.
Our users will immediately see significant improvements in speed, enhanced geocoding capability, and soon we will implement the Directions API for routing services.
Does it really make a difference what mapping provider we use? Let the numbers speak for themselves — we have seen a 10x increase in rendering speed.
During a load test of over one million Tweets and news reports across the globe, the time it took for Hyperion to render the data improved from 34 seconds with Leaflet to less than 4 seconds with Mapbox; and we expect that to improve even more as we fine-tune our vector tiles. Users will now be able to take full advantage of the massive content sets that Hyperion provides.
In addition to the increased performance on the frontend, our backend services will see a significant improvement in performance and cost savings as we scale — savings we will pass along to our clients. Chad Dalton, GeoSpark Analytics GeoArchitect, explained it like this:
“…We can serve one customer or 100,000 customers using vector tiles with the same backend resources. The load on our servers will be constant, just enough to generate vector tiles, then customers access the tiles and never reach back to our datacenter. So basically, as we add new users we don’t have to scale our database cluster to service them.”
The ongoing integration of Mapbox as our spatial engine is the first major upgrade as we transition from BlueGlass to Hyperion. Our developers were able to adopt the API quickly, and over the coming weeks, we’ll add new data sources and proprietary machine learning models which forecast political, economic, and social stability in real-time globally.
Learn more about Hyperion by Geospark Analytics. Use our business intelligence tools to add fast vector maps to your analytics platform so your users can explore hundreds of millions of data points, visualize data in real-time, and interact with custom maps. Have questions? Reach out to our team, or catch up with us at GEOINT later this month.
John Goolgasian (@JPGool) | Twitter
Geospark Analytics upgrades its maps was originally published in Points of interest on Medium, where people are continuing the conversation by highlighting and responding to this story.