User Guide

Understanding Metrics

Deep dive into health scores, performance metrics, and network analytics

Health Score System

XandScan uses a storage-first scoring system optimized for Xandeum's mission to provide exabyte-scale storage for Solana dApps.

Score Components

Storage Health (30% weight)

Optimal capacity utilization between 60-80%. Too low means wasted capacity, too high risks running out of space.

Availability (30% weight)

Percentage of nodes actively providing storage access. Higher is better.

Version Health (25% weight)

Percentage of nodes running the latest software version for compatibility.

Distribution (15% weight)

Geographic diversity and decentralization across countries and regions.

Health Score = (Storage × 30%) + (Availability × 30%) + (Version × 25%) + (Distribution × 15%)

Network Health Grade

The letter grade system provides quick assessment based on composite scoring:

A+ / A (90-100 points)

Excellent: Uptime 99%+, latency <50ms, active nodes >90%, optimal storage

B (70-89 points)

Good: Uptime 95%+, latency <100ms, active nodes 80-90%, good storage management

C (50-69 points)

Fair: Uptime 90%+, latency <200ms, monitoring recommended

D / F (<50 points)

Poor: Low uptime, high latency, or critical storage/availability issues

Performance Metrics Explained

Uptime

Percentage of time a node is online and responsive. Calculated as:

Uptime = (Time Online / Total Time) × 100

  • • 99%+ = Exceptional (less than 7 hours downtime per month)
  • • 95-99% = Very good (1-2 days downtime per month)
  • • 90-95% = Good (2-3 days downtime per month)
  • • <90% = Needs improvement

Latency

Average time for a node to respond to requests, measured in milliseconds (ms):

  • • <50ms = Excellent (local/nearby node)
  • • 50-100ms = Very good (regional node)
  • • 100-200ms = Good (distant node)
  • • 200-500ms = Fair (international node)
  • • >500ms = Poor (very distant or overloaded)

Storage Efficiency

How well storage capacity is utilized:

  • • 60-80% used = Optimal (room to grow, actively used)
  • • 40-60% used = Good (some capacity available)
  • • <40% used = Underutilized (wasted capacity)
  • • >80% used = High utilization (may need expansion soon)
  • • >90% used = Critical (expansion recommended)

Success Rate

Percentage of successful operations vs. total operations:

  • • 99%+ = Excellent (very reliable)
  • • 95-99% = Good (occasional failures)
  • • 90-95% = Fair (frequent failures)
  • • <90% = Poor (unreliable node)

Decentralization Score

Measures geographic distribution and network resilience (0-100):

  • • Based on number of unique countries and regions
  • • Considers distribution evenness (not just total countries)
  • • Higher scores indicate better fault tolerance
  • • 90+ = Highly decentralized
  • • 70-89 = Well distributed
  • • 50-69 = Moderately distributed
  • • <50 = Centralized (concentration risk)

Key Takeaways

  • • Storage and availability are weighted highest (60% combined)
  • • Health scores update every 30 seconds with live data
  • • Geographic diversity improves network resilience
  • • Version health ensures network-wide compatibility
  • • All metrics are calculated from real pNode data via pRPC