Running PTERI in production
This section explains how PTERI operates at scale, how it behaves under load, and what guarantees it provides in real-world production environments.
PTERI is designed as infrastructure, not an application. Operational simplicity and predictability are first-class goals.
PTERI verification services are stateless wherever possible.
This means:
No session affinity
No per-user server memory
No hidden state between requests
Each request contains everything needed for verification:
The message or intent
The signature
The context (challenge, address, scope)
As a result:
Requests can be handled by any instance
Horizontal scaling is trivial
Failover does not affect correctness
If a verifier goes down, another can immediately replace it.
Last updated 9 days ago
Was this helpful?
