Target Audience: Developers and SREs designing Grafana dashboards Prerequisites: SRE Golden Signals After Reading: You’ll be able to design effective dashboards and quickly identify problems

TL;DR

• Hierarchical Structure: Overview → Service → Detail order
• 5-Second Rule: Must be able to identify problem presence in 5 seconds
• Golden Signals First: Latency, Traffic, Errors, Saturation
• Remove Unnecessary Info: Exclude metrics that don’t lead to action

Why Does Dashboard Design Matter?#

Why should you invest time in dashboard layout, not just metrics? Even with perfect metrics and alerts, a poorly designed dashboard slows down incident response. If an engineer paged at 3 AM has to sift through 20 panels, it can take 10 minutes just to understand the problem. A well-designed dashboard answers “Is there a problem? Where? How severe?” within 5 seconds, dramatically reducing the time from incident detection to response.

Dashboard Design Principles#

1. 5-Second Rule#

Within 5 seconds of viewing the dashboard, you should know:

• Is there a problem?
• Where is the problem?
• How severe is it?

2. Hierarchical Structure#

graph TD
    L1["Level 1: Overview<br>Overall System Status"]
    L2["Level 2: Service<br>Individual Service Status"]
    L3["Level 3: Detail<br>Detailed Metrics"]

    L1 --> |"Problem Detection"| L2
    L2 --> |"Root Cause Analysis"| L3

This diagram shows the hierarchical drill-down structure: detecting problems at the Overview level, narrowing causes at the Service level, and deep analysis at the Detail level.

3. Color Rules#

Level 1: Overview Dashboard#

Purpose#

Grasp overall system status at a glance

Layout#

┌─────────────────────────────────────────────────────────────┐
│ Row 1: Key Metrics (Stat Panels)                             │
│ ┌────────┐ ┌────────┐ ┌────────┐ ┌────────┐ ┌────────┐      │
│ │ P99    │ │ RPS    │ │ Error  │ │ CPU    │ │ Memory │      │
│ │ 120ms  │ │ 5,234  │ │ 0.1%   │ │ 45%    │ │ 62%    │      │
│ └────────┘ └────────┘ └────────┘ └────────┘ └────────┘      │
├─────────────────────────────────────────────────────────────┤
│ Row 2: Per-Service Status (Table/Heatmap)                    │
│ ┌───────────────────────────────────────────────────────┐   │
│ │ Service    │ RPS   │ P99    │ Errors │ Status        │   │
│ │ order      │ 1,234 │ 100ms  │ 0.05%  │ 🟢            │   │
│ │ payment    │ 890   │ 250ms  │ 0.5%   │ 🟡            │   │
│ │ inventory  │ 2,100 │ 50ms   │ 0.01%  │ 🟢            │   │
│ └───────────────────────────────────────────────────────┘   │
├─────────────────────────────────────────────────────────────┤
│ Row 3: Trends (Time Series)                                  │
│ ┌───────────────────────────────────────────────────────┐   │
│ │ 📈 RPS / Error Rate Trends (24 hours)                 │   │
│ └───────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────┘

Grafana Query Examples#

# Stat: P99 Response Time
histogram_quantile(0.99, sum by (le) (rate(http_request_duration_seconds_bucket[5m])))

# Stat: Total RPS
sum(rate(http_requests_total[5m]))

# Stat: Error Rate
sum(rate(http_requests_total{status=~"5.."}[5m]))
/ sum(rate(http_requests_total[5m])) * 100

# Table: Per-Service Status
# Combine multiple queries with Transformation

Level 2: Service Dashboard#

Purpose#

Check detailed status of individual services

Layout (RED Pattern)#

┌─────────────────────────────────────────────────────────────┐
│ Variable: $service = order-service                           │
├─────────────────────────────────────────────────────────────┤
│ Row 1: Rate (Traffic)                                        │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ RPS Trend            │ │ RPS by Endpoint      │           │
│ └──────────────────────┘ └──────────────────────┘           │
├─────────────────────────────────────────────────────────────┤
│ Row 2: Errors                                                │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ Error Rate Trend     │ │ Status Code Dist.    │           │
│ └──────────────────────┘ └──────────────────────┘           │
├─────────────────────────────────────────────────────────────┤
│ Row 3: Duration (Latency)                                    │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ P50/P95/P99 Trend    │ │ Response Time Heatmap│           │
│ └──────────────────────┘ └──────────────────────┘           │
├─────────────────────────────────────────────────────────────┤
│ Row 4: Resources                                             │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ CPU Usage            │ │ Memory Usage         │           │
│ └──────────────────────┘ └──────────────────────┘           │
└─────────────────────────────────────────────────────────────┘

Variable Configuration#

# Grafana Variables
- name: service
  type: query
  query: label_values(http_requests_total, service)
  refresh: on_time_range_change

- name: instance
  type: query
  query: label_values(http_requests_total{service="$service"}, instance)

Level 3: Detail Dashboard#

Purpose#

Deep analysis for root cause identification

Layout Example (Database)#

┌─────────────────────────────────────────────────────────────┐
│ Row 1: Connection Pool                                       │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ Active Connections   │ │ Pending Connections  │           │
│ └──────────────────────┘ └──────────────────────┘           │
├─────────────────────────────────────────────────────────────┤
│ Row 2: Query Performance                                     │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ Slow Query Count     │ │ Query Execution Time │           │
│ └──────────────────────┘ └──────────────────────┘           │
├─────────────────────────────────────────────────────────────┤
│ Row 3: Resource Usage                                        │
│ ┌──────────────────────┐ ┌──────────────────────┐           │
│ │ Buffer Hit Rate      │ │ Disk I/O             │           │
│ └──────────────────────┘ └──────────────────────┘           │
└─────────────────────────────────────────────────────────────┘

Panel Type Selection#

Threshold Configuration#

Stat Panel#

{
  "thresholds": {
    "mode": "absolute",
    "steps": [
      { "color": "green", "value": null },
      { "color": "yellow", "value": 80 },
      { "color": "red", "value": 90 }
    ]
  }
}

Time Series Threshold Display#

{
  "fieldConfig": {
    "defaults": {
      "custom": {
        "thresholdsStyle": {
          "mode": "line"
        }
      },
      "thresholds": {
        "steps": [
          { "color": "green", "value": null },
          { "color": "red", "value": 500 }
        ]
      }
    }
  }
}

Best Practices#

DO (Recommended)#

✅ Dashboards with clear purpose
✅ Hierarchical drill-down structure
✅ Consistent color rules
✅ Filtering with variables
✅ Add descriptions to panels
✅ Visualization connected to alerts

DON’T (Not Recommended)#

❌ Everything in one dashboard
❌ Listing metrics without action
❌ Meaningless colors
❌ Hardcoded values
❌ Complex queries without explanation

Dashboard Sharing Configuration#

JSON Export#

# Export dashboard
curl -H "Authorization: Bearer $TOKEN" \
  "http://grafana:3000/api/dashboards/uid/abc123" \
  | jq '.dashboard' > dashboard.json

# Import dashboard
curl -X POST -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d @dashboard.json \
  "http://grafana:3000/api/dashboards/db"

Provisioning#

# grafana/provisioning/dashboards/default.yaml
apiVersion: 1
providers:
  - name: 'default'
    orgId: 1
    folder: 'Observability'
    type: file
    options:
      path: /var/lib/grafana/dashboards

Key Summary#

Next Steps#