Golden Signals Dashboard

The four golden signals of SRE,
monitored, baselined, and alerted

The Golden Signals Dashboard tracks the four metrics that matter most for service reliability: latency, traffic, errors, and saturation. Every signal is baselined against 24 hours of historical data so anomalies are detected automatically. P50, P95, and P99 latency percentiles, throughput rates, error breakdowns by type, and resource saturation across CPU, memory, and disk are all visible from a single unified view.

Start Free Trial Watch Demo
app.novaaiops.com · Golden Signals Dashboard
● LIVE
Nova AI Golden Signals Dashboard
4
Golden signals tracked
P99
Latency percentiles
24h
Baseline comparison
<30s
Anomaly detection
Latency Percentiles

P50, P95, and P99 latency: know where the slowness hides

Latency is the most user-visible signal. The Golden Signals Dashboard breaks latency into P50 (median), P95, and P99 percentiles so you can see both typical and worst-case performance. Each percentile is charted over time with a 24-hour baseline overlay, making it obvious when performance deviates from normal. Drill into any spike to see which endpoints, services, or database queries are contributing to the latency increase.

  • ✓
    Three percentile views, P50 shows typical experience, P95 shows the experience for most users, P99 catches the worst outliers
  • ✓
    24-hour baseline overlay, a shaded band shows yesterday's pattern so you can distinguish normal daily patterns from true anomalies
  • ✓
    Endpoint-level drill-down, click any latency spike to see which specific endpoints and downstream calls are responsible
app.novaaiops.com · Golden Signals · Latency
Latency percentiles P50 P95 P99
Traffic & Error Rates

Throughput trends and error breakdowns in real time

Traffic throughput is displayed as requests per second with breakdowns by service, endpoint, and HTTP method. Error rates are shown as both absolute counts and percentages, categorized by error type (4xx client, 5xx server, timeout, connection refused). The dashboard correlates traffic spikes with error rate changes so you can distinguish between load-induced errors and functional bugs.

  • ✓
    Requests per second, real-time throughput with breakdown by service, endpoint, and HTTP method
  • ✓
    Error categorization, errors are grouped by type: 4xx, 5xx, timeouts, connection failures, and custom error codes
  • ✓
    Traffic-error correlation, visual overlay shows when error rate increases correlate with traffic spikes vs. independent failures
app.novaaiops.com · Golden Signals · Traffic & Errors
Traffic throughput and error rate charts
Resource Saturation

CPU, memory, and disk: catch saturation before it causes outages

Saturation measures how close your resources are to their limits. The dashboard tracks CPU utilization, memory pressure, disk I/O, and network bandwidth across all services and hosts. Each metric is displayed with threshold lines for warning (70%) and critical (90%) levels. Predictive trendlines estimate when resources will hit capacity so you can scale proactively rather than reactively.

  • ✓
    Multi-resource view, CPU, memory, disk I/O, and network bandwidth all tracked with consistent threshold indicators
  • ✓
    Predictive capacity planning, trendlines forecast when resources will hit warning and critical thresholds based on current growth
  • ✓
    Per-service breakdown, see which services are consuming the most resources and identify candidates for optimization
app.novaaiops.com · Golden Signals · Saturation
Resource saturation monitoring
Video walkthrough coming soon

Subscribe to Nova AI Ops on YouTube for demos, tutorials, and feature deep-dives.

Experience Golden Signals Dashboard in action

Golden Signals Dashboard is part of the Nova AI Ops platform. Start a free trial to see it with your own data.

Start Free Trial Request a Demo