Resilient, high-available and pooled parallel Connections

Resilient and high-available connections are established through redundant endpoints and transparent failover, while a pool of concurrent persistent connections ensures low-latency parallel request handling and avoids classical head-of-the-line blocking.

• Reconnection logic
  • Initial waiting time
  • Backoff
• H/A settings
  • DNS-SRV records
  • HTTP 429 (Too many requests)
  • Static configuration within the device model
• Extended DNS-SRV records
  • Server public keys for security
  • Information how many concurrent connections are allowed for connection pooling
• Concurrent Connection Pools
  • Shall some connections be dedicated to certain protocol functions, e.g. RFID Auth/RemoteStarts within EV roaming over a dedicated connection pool for lowest latency and to avoid head-of-the-line blockings?