District Maps

Mapping Landscape Architecture

District Maps, serving landscape architects and urban planners, faced slow performance limiting project scale. Ollon restructured the system to handle large geographic datasets. Planners now collaborate in real time on complex projects, navigate maps seamlessly, and visualize development proposals without performance delays that previously hindered their work.

Situation

District Maps, a mapping platform for landscape architects and urban planners, needed a scalable, web based GIS system to manage large geographic datasets and collaborative mapping projects. Their initial development team struggled with performance issues, slow map rendering, and inefficient data processing, making the platform difficult to use for large scale projects. They required a solution that would optimize performance, enhance user experience, and support real time collaboration.

Solution

Ollon restructured the GIS architecture, improving map rendering speeds, data processing efficiency, and overall performance. A hybrid approach using dynamically drawn map layers and static vector tiles was implemented to ensure smooth navigation and responsiveness, even with large datasets. Additionally, collaborative tools were developed, allowing multiple users to edit and visualize mapping projects in real time, streamlining workflows for landscape architects and planners.

Results

The revamped system provided better performance, improved scalability, and enhanced collaborative data manipulation for landscape architects. District Maps now supports larger scale mapping projects with real time updates and improved usability.

Technologies

Python, Django, MapBox, PostgreSQL, AWS

Our Expertise in Action

Application Triage and Rescue

District Maps, a platform landscape architects and urban planners use to manage large geographic datasets, was struggling under heavy processing and slow rendering, particularly when uploading map data they already had in a file. Ollon's rescue work addressed the failing build on multiple fronts: separating large uploads onto their own temporary infrastructure, restructuring the Django backend that moves and processes the data, and introducing a hybrid map rendering approach combining dynamic layers with static vector tiles. The platform now takes on far larger projects than the original build could support, without the slowdowns that used to affect every user on it.

Fractional DevOps Support

District Maps, a mapping platform, ran day to day without a dedicated DevOps hire, leaving server configuration, security rules, and deployment to be set up correctly once and maintained going forward. Terraform templates, adapted from Ollon's own internal setup, configure exactly which IPs and VPN connections are allowed to reach District Maps, and an automatic build pipeline deploys the staging site whenever a commit lands on the front end or back end staging branch. Over roughly two years, DevOps work has remained a small fraction of the project's total development effort, evidence that the infrastructure runs largely on its own.

GIS/Mapping

Landscape architects and urban planners use District Maps to build and collaborate on maps across large geographic datasets, drawing shapes, lines, and layers the way a designer would work in Photoshop. Ollon combined dynamically drawn map layers with static vector tiles in a hybrid rendering approach, keeping the platform responsive as project data grows in size and complexity. Built on MapBox, that architecture supports real time collaboration, letting multiple planners edit and visualize the same mapping project at once. Planners on the same project now see each other's changes as they happen, instead of working from a version of the map that could already be out of date.

Django

Before this restructuring, District Maps processed and moved geographic data slowly enough that large projects triggered the rendering and performance problems planners were running into. Ollon used the Django framework to rework how the application handled those large geographic datasets, changing how data moved through the system and how quickly maps rendered for users. Paired with PostgreSQL for storage, that restructuring runs underneath the platform's visible speed improvements, and it is what lets District Maps keep pace as project sizes continue to grow.

AWS

Large geographic data uploads to District Maps used to compete with everything else running through its browser and shared backend. AWS instances now spin up on demand to process a user's file upload, then shut down once the work is done. When several users upload large files at the same time, each gets its own instance instead of waiting behind a shared queue. That isolation is what lets one user's upload run independently of everyone else's, with AWS costs rising only when someone is actually processing a file.