SYSTEM_METRICS
Monitor infrastructure performance, latency, and error rates
[ TOTAL_QPS ]
104.7
Queries per second across all routes
[ AVG_ERROR_RATE ]
2.53%
Average across all routes
[ AVG_LATENCY ]
451ms
Average response time
ROUTES_PERFORMANCE
Detailed metrics for each API endpoint
| Route | QPS | Avg Latency | Error Rate |
|---|---|---|---|
| /api/classify | 27.7 | 367ms | 2.50% |
| /api/qa | 26.8 | 515ms | 2.39% |
| /api/extract | 25.7 | 479ms | 2.86% |
| /api/summarize | 24.5 | 444ms | 2.37% |
ERROR_RATE_BY_ROUTE
Lower is better - fewer failures
LATENCY_BY_ROUTE
Lower is better - faster response times
THROUGHPUT_BY_ROUTE
Queries per second for each endpoint
SYSTEM_HEALTH_INSIGHTS
Highest Traffic Route
Route:/api/classify
QPS:27.7
Latency:367ms
Slowest Route
Route:/api/qa
Latency:515ms
QPS:26.8
Most Errors
Route:/api/extract
Error Rate:2.86%
Status:Warning
System Recommendations
- Consider scaling for high traffic
- Monitor trends over longer time periods
SYSTEM_UPTIME
Service availability and status monitoring
API GATEWAY
99.9%
24h uptime
MODEL SERVICE
99.8%
24h uptime
DATABASE
100%
24h uptime
ANALYTICS
99.7%
24h uptime
LAST 30 DAYS
ORANGE = UP • GRAY = INCIDENTS
RECENT INCIDENTS
Model Service Timeout
Nov 28, 2024 14:23 UTC • Resolved in 4m
2.1% downtime
API Gateway Latency Spike
Nov 25, 2024 09:15 UTC • Auto-resolved in 12m
Performance degraded
Scheduled maintenance completed
Nov 20, 2024 02:00 UTC • 30m planned downtime
Maintenance
METRIC_DEFINITIONS
QPS (Queries Per Second)
Number of requests handled per second. Higher values indicate more traffic. Use for capacity planning.
Average Latency
Mean response time for requests. Lower is better. Spikes indicate performance degradation or bottlenecks.
Error Rate
Percentage of failed requests. Target <2% for production systems. Investigate routes with >5% error rate.