Are there recommended infrastructure sizing guidelines (CPU, RAM, storage, node/replica counts) for Docker or Kubernetes deployments based on...

Partial guidance emerged on July 20, 2026. Paysyslabs shared indicative specs based on their own UAT environment for a three-server Docker deployment: server-a (4 cores / 16GB RAM / 50GB disk, core services including TMS), server-b (4 cores / 16GB RAM / 50GB disk) and server-c (8 cores / 64GB RAM / 150GB disk), the latter two running the Release 4.0 extensions - see the tazama-stack aws-deployment-instructions.md for the full breakdown of which services run on each server. Paysyslabs was clear these are UAT reference specs only, not a validated TPS benchmark - no formal throughput/performance analysis has been done yet. Caveat: these figures describe an entry-level Docker Compose deployment, not a production-grade Kubernetes substrate, and should not be used to correlate or extrapolate high-end performance; validated sizing guidance requires formal benchmarking on production-grade Kubernetes infrastructure, which has not yet been performed. A related question about guaranteeing 50ms response times was also clarified: the team is not targeting 50ms internally for the 4.0 release - the current internal target is 200ms. The 50ms figure some community members had seen came from a Release 3.0.0 YouTube demo and does not reflect the 4.0 target. A broader performance evaluation of Release 4.0 is planned but not yet scheduled. If you have production sizing experience of your own, consider sharing it in #get-help while comprehensive official guidance is pending.