The World Bank needed to deliver accurate donor-funded project information to stakeholders across multiple continents without overprovisioning global infrastructure. ASCENDING designed and implemented a serverless AWS architecture that improved response time for remote users, reduced network errors, and lowered cost through pay-per-use scaling.
Background
The World Bank operates across 189 member countries with offices in more than 130 locations. Its teams manage thousands of development projects, each with strict procedural and reporting requirements designed to protect project integrity and support aid delivery.
Because project data had to be accessible worldwide, the platform needed to support fluctuating demand from users in North America, Europe, Asia, and Africa. The existing architecture could not consistently provide low-latency access for global stakeholders.
The Challenge
The World Bank faced a tradeoff between performance and efficiency. Visitors outside the United States frequently experienced delayed responses or network errors due to limited endpoint coverage, while average infrastructure utilization remained low for much of the day.
- Global users needed faster and more reliable access to project information.
- Traffic patterns were inconsistent, making fixed server capacity inefficient.
- Dedicated always-on infrastructure in multiple regions would create unnecessary cost.
- Internal security controls required access restrictions for credential-sensitive workloads.
- Heavy compute tasks could not run directly inside Lambda resource limits.
Why ASCENDING
ASCENDING was selected based on practical delivery experience in AWS serverless architecture and infrastructure automation for organizations with different operating scales.
The team also demonstrated a delivery model that balanced technical depth with responsive communication, helping stakeholders align on architecture decisions, security constraints, and rollout execution.
The Solution
ASCENDING implemented an event-driven serverless architecture with AWS Lambda and Amazon API Gateway as the global request interface, while offloading higher-compute workloads to Amazon EC2 and AWS Machine Learning services.
- Client requests were received through Amazon API Gateway.
- API Gateway invoked dedicated AWS Lambda functions for create, update, delete, and lookup operations.
- Lambda functions performed orchestration and routed compute-intensive tasks to Amazon EC2 and AWS Machine Learning components.
- Lambda collected processing outcomes and returned responses through API Gateway.
- Security controls were added, including employee authorization checks, IP restrictions, and additional validation logic for protected internal use cases.
This architecture provided elasticity for variable global usage while preserving cost efficiency through consumption-based execution.
The Outcome
The implemented solution improved global delivery performance and operational visibility while reducing ongoing cloud spend.
- 70% faster response for remote visitors accessing project information.
- 90% reduction in observed network errors for international users.
- 50% lower Lambda cost according to the latest post-implementation billing review.
- More standardized deployment workflows for application updates.
- Better visibility into cloud resource consumption, traffic behavior, and budget planning.
- Improved security posture for internal projects requiring controlled access.
