Customer Stories
Scaling Smarter: How USGS-WDFN Transformed Its Environmental Data Platform
Problem
The U.S. Geological Survey’s Water Data for the Nation (USGS-WDFN) operates a high-volume, public-facing environmental data platform built on open geospatial standards. As traffic grew, limitations in application concurrency and database connection management forced inefficient scaling driving rising cloud costs and increasing operational risk during peak demand periods.
Solution
eSimplicity re-engineered how the platform handled concurrent requests and database interactions, aligning application-level behavior with real user demand. By tuning Asynchronous Server Gateway Interface (ASGI) workers, thread pools, connection pooling, and container auto-scaling triggers, the team dramatically increased per-container throughput while preserving reliability and standards-based API access.
Outcomes
The modernized architecture reduced infrastructure requirements and costs by more than 92% while reliably supporting peak traffic of four million requests per day. USGS gained predictable performance, substantial cost savings, and a scalable technical foundation for continued growth of public hydrologic and geospatial data services.
MEASURABLE IMPACT
We produce results for our customers
USGS-WDFN delivers mission-critical hydrologic and geospatial data through a pygeoapi-based service implementing Open Geospatial Consortium (OGC) standards. The platform must support scientists, application developers, and downstream systems that rely on consistent, low-latency access to authoritative environmental data.
As usage increased, the system required a large number of Elastic Container Service (ECS) containers to handle production traffic, even though individual containers were significantly underutilized. Concurrency limits related to ASGI worker configuration, thread pools, and database connection pooling constrained throughput, forcing horizontal scaling as the only way to absorb demand.
This approach increased cloud costs and introduced operational risk. Scaling decisions were driven by coarse infrastructure metrics rather than actual request concurrency, making it harder to predict behavior under peak load and limiting confidence in the platform’s ability to scale sustainably.
Process & Solution
eSimplicity engineers conducted a focused, application-level optimization effort to address the root causes of inefficient scaling rather than masking them with additional infrastructure.
The team tuned ASGI worker counts and thread pool behavior to increase effective request concurrency per container, optimized SQLAlchemy connection pooling to remove database bottlenecks, and redesigned ECS auto-scaling logic to respond to active load balancer connections instead of CPU or memory utilization alone.
These changes ensured that scaling decisions reflected real user demand and that each container delivered substantially more useful work. The result was a more stable, predictable system capable of sustaining high request volumes without unnecessary over-provisioning.
Outcomes
The optimized architecture delivered major gains in efficiency, stability, and cost effectiveness. The platform now scales smoothly under sustained peak traffic, reducing operational risk while significantly lowering cloud spend. By aligning scaling behavior with real request concurrency, greater confidence was gained in system performance during high-demand events and seasonal usage spikes.
Beyond immediate cost and performance improvements, eSimplicity’s work established a more sustainable and transparent operating model. The modernized platform enables predictable growth without proportional infrastructure increases, ensuring reliable, low-latency access to mission-critical hydrologic and geospatial data. As a result, stakeholders can be better supported with a foundation for expanding data services in the future.
