Major Infrastructure Overhaul: Scaling for the Future

At Thirdweb, we’ve been hard at work implementing significant improvements to our infrastructure, and we’re thrilled to share the results of these efforts. This update represents a major leap forward in scalability, performance, and resilience, ensuring we’re better equipped to serve our users now and in the future. Here's what we’ve achieved:

⚡ Unlocking New Levels of Performance

One of the key drivers of this overhaul was the need to reduce latency and improve response times. Our engineering team has:

• Reduced P99 Latency: We’ve slashed P99 latency from upwards of 20 seconds to just 2 seconds. This improvement means even the slowest requests now complete faster than ever.
• Optimized Average Response Time: The average response time is now just 250ms, providing a snappier, smoother experience for our users.
• Expanded Regional Coverage in Europe and Asia Pacific: We have introduced new regions in Europe and Asia Pacific, allowing us to provide more localized services to meet the growing demands of our global customer base. Additionally, these new regions enhance our disaster recovery capabilities, ensuring greater resilience and reliability.

These enhancements ensure that our platform is not only faster but also more consistent under varying workloads.

📈 Enhanced Autoscaling for Seamless Scaling

To handle ever-increasing traffic demands, we’ve rolled out improved autoscaling capabilities. These dynamic systems allow us to:

• Respond to peak load surges without sacrificing performance.
• Maintain cost efficiency by scaling resources up or down as needed.

This ensures that no matter how much demand our platform faces, we’re ready to deliver resilient service.

🛡️ Stress-Tested for Resilience

Reliability is critical, especially as we scale. To validate our infrastructure’s ability to handle increasing demand, we conducted resilience testing at over 6,000 Requests Per Second (RPS). The system performed admirably, solidifying confidence in its ability to handle extreme loads.

Our load testing focused on the RPC router, which efficiently routed requests to over 2,000 nodes globally over a 60-minute period. The screenshot below provides a sample measurement of the aggregated P90 latency across all RPC calls.

🌍 Regional RPC Rollout

We’re expanding our global footprint with the rollout of Regional RPC services in Asia Pacific (SEA), Europe (EU), and Americas (US). This regionalization improves latency by up to 60% for users across the globe and ensures a more localized and reliable service experience.

🔐 Updated Authentication with Team Integration

Security and collaboration are critical, and our updated authentication system now integrates seamlessly with teams. This allows:

• Enhanced access control for better security.
• Simplified management of team-based workflows.

This update ensures that working as a team on our platform is both secure and effortless.

🔮 Looking Ahead

These improvements mark a significant milestone in our ongoing commitment to delivering a fast, reliable, and scalable platform for our users. But we’re not stopping here. Our team is already planning the next wave of enhancements to stay ahead of the curve and continue providing the best experience possible.

Thank you for being a part of our journey. We’re excited about these changes and can’t wait for you to experience them. Let us know your thoughts and stay tuned for more updates!