Optimising IoT load testing with Hoverfly Cloud

How an animal health organisation optimised IoT load testing

Imagine managing a global network of over 20,000 IoT devices, each transmitting critical animal health data in real-time. This is the reality for a leading animal health organisation. But with such massive scale comes a daunting challenge: how to ensure the reliability and performance of their cloud infrastructure under extreme load?

The organisation turned to Hoverfly Cloud to enhance their load testing capabilities.

The challenge: Testing at scale

The organisation's livestock monitoring systems include over 20,000 IoT devices deployed globally to monitor animal health. These devices send telemetry data to cloud-based data stores for near-real-time reporting, ensuring secure data transfer via an Identity and Access Management (IAM) system.

To validate the cloud infrastructure's ability to handle massive loads, performance engineers needed to simulate high volumes of IoT authentication requests. However, using the real IAM system for this load testing introduced several challenges:

• Operational complexity: Maintaining 20,000+ test accounts for authentication would be time-consuming and error prone.
• High costs: Real IAM systems incur substantial costs when scaled for non-production environments.
• Focus shift: Testing the IAM system, not the cloud infrastructure, became an unintended distraction.

The organisation required a solution to decouple IAM from the performance testing process.

The solution: Hoverfly Cloud in action

Engineers implemented Hoverfly Cloud to simulate the organisation’s IAM services.

• Token issuance: Hoverfly simulates the OIDC provider token endpoint and provides a pre-created valid access token.
• Token validation: Hoverfly simulates the authentication server endpoint that an application can use to retrieve the public key to validate the token.

This setup allowed the organisation to focus exclusively on their AWS-hosted cloud infrastructure’s performance under extraordinary loads. Hoverfly Cloud provided the flexibility and reliability needed to replicate production-like conditions without real IAM dependencies.

Results and benefits

Hoverfly Cloud delivered significant advantages:

1. Cost savings
   • Eliminated the need to maintain 20,000+ test accounts, reducing operational expenses.
   • Avoided licensing and resource costs associated with a full-scale test IAM system.
2. Simplified management
   • Reduced setup time for load tests by replacing a complex IAM dependency with a lightweight simulation.
   • Enabled faster iterations with predictable API responses.
3. Control of APIs
   • Provided a deterministic environment for token issuance and validation.
   • Ensured consistent test conditions, boosting the reliability of performance metrics.

Key metrics:

• Setup time: Reduced by over 50%.
• Scalability: Successfully simulated 1 million+ requests per hour without degradation.

With Hoverfly Cloud’s API simulation, you can significantly enhance your testing strategy, improve application quality, and accelerate time-to-market

Conclusion

By adopting Hoverfly Cloud, the organisation eliminated the operational complexity and costs associated with integrating the IAM into their load tests. This strategic decision enabled performance engineers to focus entirely on optimising cloud infrastructure, ensuring robust and reliable IoT data flows at scale.

Hoverfly Cloud has proven to be an essential tool in the organisation’s performance engineering toolkit, setting a benchmark for efficient and cost-effective load testing.

See Hoverfly Cloud in action