Performance Benchmarks

📄️ Latency Footprint

Through our benchmarking sessions, we wanted to unveil Lunar's latency footprint on response time percentiles in comparison to direct API calls made without Lunar's intervention. In our experiments, we have selected a provider with a constant response time of 150ms, which is relatively fast for a web-based API.

📄️ Capacity Benchmark

In our benchmarking sessions, we evaluated Lunar's capacity to handle requests per second under various loads. Using an EKS cluster with a reliable hardware configuration, we generated traffic to measure performance. By adjusting parameters like concurrent connections and request rates, we aimed to assess Lunar’s efficiency, identify bottlenecks, and pinpoint areas for improvement in handling large workloads.

In today's API-driven world, swift response times are crucial for both providers and consumers. At Lunar, we understand the importance of minimizing latency and have designed our solution to bridge the gap between API providers and consumers without adding significant overhead.

After extensive research and benchmarking, a single instance of Lunar introduces only 4ms latency at the 99th percentile while handling up to 84,867 requests per second. This ensures optimal performance and minimal impact on overall system latency.

More details about our Latency Footprint and Capacity Benchmark can be found in the sub-pages below.