Sustainability Dashboard

Real-time monitoring and analytics for your AI's environmental impact.

Dashboard Overview

The Sustainability Dashboard provides comprehensive insights into your carbon footprint and efficiency improvements.

Key Metrics

Carbon Footprint Tracking

• Total Emissions: Overall carbon impact of your AI usage
• Emissions by Provider: Breakdown by AI provider
• Emissions by Model: Detailed model-level analysis
• Time Series: Historical trends and patterns

Carbon Savings

• Total Savings: Carbon reduced through GateFlow optimization
• Sustain Mode Impact: Savings from automatic routing
• Caching Impact: Emissions avoided through semantic caching
• Time-Shift Savings: Carbon reduced through deferred execution

Accessing the Dashboard

1. Login to your GateFlow account
2. Navigate to Analytics → Sustainability Dashboard
3. Select your time range and filters

Dashboard Sections

Summary View

# Get sustainability summary via API
summary = client.sustainability.summary.get()

print(f"Total carbon footprint: {summary.total_carbon_gco2e} gCO₂e")
print(f"Carbon saved: {summary.total_carbon_saved_gco2e} gCO₂e")
print(f"Savings percentage: {summary.savings_percentage}%")
print(f"Requests optimized: {summary.requests_optimized}")
print(f"Current carbon intensity: {summary.current_grid_intensity} gCO₂/kWh")

Time Series Analysis

# Get historical data
time_series = client.sustainability.time_series.get(
    start_date="2024-01-01",
    end_date="2024-01-31",
    granularity="daily"
)

for day in time_series.results:
    print(f"{day.date}: {day.carbon_gco2e} gCO₂e (saved: {day.carbon_saved_gco2e})")

Provider Comparison

# Compare provider efficiency
provider_comparison = client.sustainability.providers.compare()

for provider in provider_comparison.providers:
    print(f"{provider.name}:")
    print(f"  Requests: {provider.request_count}")
    print(f"  Avg carbon/request: {provider.avg_carbon_per_request_gco2e} gCO₂e")
    print(f"  Efficiency score: {provider.efficiency_score}/100")

Dashboard Features

Real-time Monitoring

• Live carbon intensity maps
• Current grid conditions
• Active request carbon impact

Alerts and Notifications

• High carbon intensity warnings
• Optimization opportunities
• Provider outage alerts

Export and Reporting

• CSV/Excel export for ESG reporting
• PDF reports for stakeholders
• API access for integration

Using Dashboard Data for ESG Reporting

Key Metrics for Reports

1. Scope 3 Emissions: AI-related carbon footprint
2. Reduction Initiatives: Sustain Mode adoption and impact
3. Efficiency Improvements: Year-over-year carbon intensity reduction
4. Provider Diversity: Multi-provider strategy for resilience

Report Template

# AI Sustainability Report - Q1 2024

## Executive Summary
- **Total AI Carbon Footprint**: 1,250 kg CO₂e
- **Carbon Savings**: 875 kg CO₂e (41% reduction)
- **Sustain Mode Adoption**: 78% of eligible requests

## Detailed Metrics

### Carbon Footprint Breakdown
- **Compute**: 850 kg CO₂e
- **Data Transfer**: 200 kg CO₂e  
- **Storage**: 200 kg CO₂e

### Savings by Optimization
- **Sustain Mode**: 550 kg CO₂e
- **Semantic Caching**: 200 kg CO₂e
- **Time-Shifting**: 125 kg CO₂e

### Provider Efficiency
| Provider | Requests | Carbon/Request | Efficiency Score |
|----------|---------|---------------|------------------|
| Cohere | 12,500 | 45 gCO₂e | 92/100 |
| Anthropic | 8,200 | 60 gCO₂e | 88/100 |
| OpenAI | 5,100 | 75 gCO₂e | 85/100 |

Advanced Analytics

Carbon Intensity Heatmap

# Get grid carbon intensity data
heatmap_data = client.sustainability.grid_intensity.get(
    region="north_america",
    time_range="last_7_days"
)

# Visualize high/low carbon periods for optimal scheduling

Model Efficiency Leaderboard

# Get most efficient models
efficiency_ranking = client.sustainability.models.rank()

for model in efficiency_ranking.top_10:
    print(f"{model.rank}. {model.name} ({model.provider}): {model.efficiency_score}")

Dashboard Configuration

Custom Views

Create personalized dashboards for different teams:

1. Executive View: High-level metrics and trends
2. Developer View: Technical details and optimization opportunities
3. Compliance View: ESG reporting data and audit trails

Alert Thresholds

Set custom alert thresholds:

# Configure alerts
client.sustainability.alerts.configure(
    high_carbon_threshold=150,  # gCO₂e per request
    optimization_opportunity_threshold=25,  # % potential savings
    provider_outage_notification=True
)

Best Practices for Dashboard Usage

Regular Monitoring

• Daily: Check current carbon intensity and active optimizations
• Weekly: Review provider performance and efficiency trends
• Monthly: Generate reports and identify improvement opportunities

Team Collaboration

• Share dashboard access with relevant teams
• Set up automated reports for stakeholders
• Use dashboard data in sprint planning and architecture reviews

Continuous Improvement

• Identify high-impact optimization opportunities
• Track progress toward sustainability goals
• Celebrate milestones and share successes

Troubleshooting

"Dashboard data not updating"

Solutions:

1. Check your API key permissions
2. Verify request logging is enabled
3. Ensure Sustain Mode is properly configured
4. Contact support if issues persist

"Carbon savings seem low"

Solutions:

1. Review your quality thresholds in Sustain Mode
2. Check grid carbon intensity in your region
3. Add more providers for better optimization options
4. Enable additional features like time-shifting and caching

Next Steps

• Explore Sustain Mode - Deep dive into automatic optimization
• Configure Alerts - Set up custom notifications
• Export ESG Data - Generate compliance reports
• View Provider Analytics - Compare provider efficiency