Metrics
The metrics dashboard has several views:Overall
The top-level view showing Total RPC Requests across all networks, with a time-series sparkline showing request volume trends.
Usage
- Network Usage: Time-series chart showing request volume broken down by blockchain network
- Networks Share: Pie chart displaying the distribution of traffic across different chains (e.g., plasma-mainnet, monad, base, swell-mainnet)
- Method Usage: Time-series showing which RPC methods are being called over time
- Methods Share: Pie chart breakdown of method distribution (e.g.,
eth_call,eth_getBalance,eth_getTransactionByHash)
Networks
- Network-level Critical Errors: Track critical failures by network and error type
- Network-level Warnings: Monitor warnings like missing data from endpoints
- Network-level Notices: Informational events such as upstream request issues
- Incoming Requests: Request volume per network over time
- Incoming RPS: Requests per second breakdown by network
- Successful Responses: Success rates with empty vs non-empty response tracking
- Network-level P99/P90/P50 Resp. time: Response time percentiles by network for performance analysis
- Multiplexed requests: Requests that were deduplicated across multiple callers
- Response Finality Share: Distribution of finalized vs real-time responses
- Rate-limited requests: Self-imposed rate limiting events
- Hedge Effectiveness: Performance of hedging mechanisms across providers
