Inspiration: Scaling microservices often feels like a game of guess word. We built Velocity Mesh to turn the chaos of high-traffic growth into a precise, visual craft.
What it does: It is a lightweight monitoring and optimization layer that visualizes data flow and suggests real-time configuration changes to prevent bottlenecks.
How we built it: The core was built using Go and gRPC for high-speed networking. We used a basic load-efficiency formula to calculate the performance coefficient-P=U(avg)/L(max)
Challenges we ran into: Reducing the "mesh tax" (added latency) was our biggest hurdle. We had to heavily optimize our middleware to ensure monitoring didn't slow down the actual production traffic.
Accomplishments that we're proud of: Our Auto-Scale Suggester successfully predicted and flagged potential service failures before they actually happened during our stress tests.
What we learned: We gained deep insights into service mesh observability and the importance of prioritizing core stability over adding too many secondary features.
What's next for Velocity Mesh: Automating the suggested fixes so the system can self-heal without manual intervention and adding mTLS for better security between nodes.
Log in or sign up for Devpost to join the conversation.