Performance Characteristics

Measured and estimated performance characteristics for TameFlare components. All numbers are from local testing on commodity hardware (4-core CPU, 16 GB RAM, SSD).

Note
TameFlare has not been independently benchmarked. These numbers are internal measurements and should be validated in your environment.

Proxy latency (Gateway)

The gateway adds latency to every proxied request. This is the overhead TameFlare introduces on top of the upstream API's response time.

MetricHTTPHTTPS (with TLS interception)
p50~1-2ms~3-5ms
p95~3-5ms~8-12ms
p99~8-15ms~15-25ms

What contributes to latency

StepTimeNotes
Connector lookup (domain → connector)<0.1msIn-memory map
Action parsing (HTTP → structured action)~0.5-1msRegex matching on URL/method
Permission check (SQLite lookup)~0.5-2msCached after first lookup
Credential injection<0.1msIn-memory vault
TLS to upstream API~2-5msFresh TLS handshake to upstream; session resumption after first request
Traffic log write (SQLite)~0.5-1msAsync, non-blocking

First request to a new domain

The first HTTPS request to a new domain takes ~20-50ms extra because the cloud gateway performs connector lookup, permission resolution, and a fresh TLS handshake to the upstream API. Subsequent requests to the same domain benefit from cached configuration and TLS session resumption.


Gateway resource usage

MetricIdleUnder load (100 req/s)
RAM~20-30 MB~40-80 MB
CPU<1%~5-15% (4-core)
Goroutines~10~200-500
Open file descriptors~20~300-500

The gateway is a single Go binary with no external dependencies. Memory usage scales linearly with concurrent connections.


Cold start

ComponentTimeNotes
Cloud gateway~50-200msConfig fetch + vault decrypt + CA load
First proxied request+10-50msConnector init + first TLS cert generation

Scaling limits

DimensionPractical limitBottleneck
Concurrent agents~50-100 per gatewayPort allocation (one port per agent)
Requests per second~500-1,000 per gatewaySQLite write throughput for traffic logs
Policies~100-200Evaluation time grows linearly
Audit events~10M rows before queries slowSQLite full-table scan. Use AUDIT_RETENTION_DAYS to prune.
Traffic log size~1 GB per ~5M requestsDisk space. Prune via maintenance endpoint.

For higher throughput, create multiple gateways in the dashboard and run them on separate machines.


Next steps