wtf is execution-payloads-benchmarks?

Execution Payloads Benchmarks is a testing tool that measures how well Ethereum execution clients handle realistic workloads. If you're building or evaluating blockchain node software and need to know how it performs under stress, this tool automates the process of spinning up clients, blasting them with transactions, and collecting detailed performance data. At a high level, you write a configuration file describing what you want to test, which client, which scenarios, how long, and so on. The tool then uses Docker to spin up an isolated environment, runs the scenario against the execution client, and uses Grafana K6 (a popular load-testing tool) to generate and fire transaction payloads at the client. After the run, it dumps everything into an output directory: client logs, K6 test scripts, performance summaries, and raw metrics. The target users are blockchain engineers and researchers who need hard numbers on how different execution clients behave under load. For example, if your team is optimizing a client for high-throughput environments, you'd run a scenario that simulates a busy block and compare processing times across versions. Or if you're evaluating which client to deploy in production, you could run multiple scenarios in a loop to see which one holds up best over sustained pressure. A notable aspect of the project is its heavy use of the Grafana ecosystem for observability. Beyond writing results to local JSON files, it can stream live metrics to Prometheus and send profiling data to Grafana Pyroscope, which lets you visualize what's happening inside the client in real time. This makes it less of a simple benchmark script and more of a full performance lab setup. The README notes that Pyroscope integration is still a work in progress, so some features may not be fully polished yet.