Mapping tools benefit enterprises across industries like logistics, retail, and real estate. They are essential for route optimization, fleet management, and delivery tracking. Retailers use mapping tools to choose optimal store locations, analyze customer demographics, and manage supply chains. They also help real estate companies assess property values, analyze market trends, and identify development opportunities.
Table of Contents
- Assessing the scalability of route mapping tools
- Calculating the shortest path between nodes in one segment
Key Takeaways
- Google Maps is popular, but its capabilities are limited
- Route mapping tools should account for construction, accidents, and traffic volume
- Enterprise-level mapping tools incorporate billions of edges and nodes
- Loading all edges and nodes into memory simultaneously is not feasible
- The graph is divided into segments to remain scalable
- You calculate the weight between the edges linking nodes
As of August 2024, Google Maps had more than a billion monthly active users. In the US, the platform attracts over 158 million unique users per month. It generates more than 69,000 searches per second and has been downloaded over 5 billion times on Google Play.
Despite those numbers, Google Maps has real limitations at the enterprise level. Consider exterior lighting management as an example. Some companies use Google Maps to document fixture locations and types across their contracted properties. Those maps need to be shared with clients, and that is where things break down. Google Maps works well as a free consumer tool, but sharing maps in formats like PDF is cumbersome. Clients need a Gmail account just to view a shared map.
The best maps software for this kind of enterprise application would let you add legends, distinguish fixture types with different colors or shapes, and export maps as a PDF or another standard format. Clients should be able to view what they need without creating an account.
Assessing the scalability of route mapping tools
Route mapping is another area where enterprise demands push well past what standard tools can handle. Companies like Lyft, Uber, and Waymo use these tools internally, and the underlying maps involve billions of edges and nodes when calculating routes across the globe. Running Dijkstra or another standard pathfinding algorithm on a network that large produces a poor user experience. It simply does not scale. A modified approach is required.
ETA calculations also need to account for road construction, accidents, and traffic volume, updating dynamically as conditions change. Traffic volume is consistently rising. Travel on all US streets and roads increased by 4.8 billion vehicle miles in February 2024 year over year, for a total of 240.2 billion vehicle miles.
Enterprise-level mapping tools handle a network of billions of edges and nodes. Running a shortest-path algorithm directly on the full graph is not practical. Loading every edge and node into memory at once is not possible. To remain scalable, the graph is broken into small segments, producing smaller polygons where calculating the shortest path becomes manageable. Performance improves further by querying multiple polygons in parallel.
Calculating the shortest path between nodes in one segment
Within each segment, there is a weight assigned to the edges connecting two nodes. That weight is calculated based on factors like average vehicle speed and traffic density. The segment is then saved, and a standard shortest-path algorithm can run against it, with results cached for future use. This approach scales to a degree because only the segments relevant to a given user request are loaded. If segment data volume increases, multiple servers become necessary. Relying on a single server as volume grows will cause performance to deteriorate.
FAQ
What are the metrics of a scalable system?
A scalable system has three standard metrics: resource use (memory and CPU), transactions per second, and response time.
How does mapping technology work?
Mapping technology uses high-resolution cameras, satellites, and computer software to collect data and generate maps. Cloud and web-based solutions like Maptive require no downloads. You always have access to the most current data from any device.
