// Free Developer Tool

Execution Latency Calculator.

Configure your bot's current setup and see exactly where milliseconds are lost. Compare RPC polling, WebSocket, raw gRPC, and pre-parsed gRPC side by side.

Your current setup

100ms5000ms
1ms (co-located)100ms (remote)
5ms (Rust/optimized)50ms (JS/unoptimized)
1ms50ms

Execution latency comparison

JSON-RPC Polling548ms
Data: 515msParse: 28msLogic: 5ms
WebSocket128ms
Data: 95msParse: 28msLogic: 5ms
Raw gRPC (Yellowstone)53ms
Data: 20msParse: 28msLogic: 5ms
Subglow gRPC (pre-parsed)25ms
Data: 20msParse: 0msLogic: 5ms
523ms faster

Switching from JSON-RPC Polling to Subglow saves 523ms per transaction. 28ms of that is parsing overhead that Subglow eliminates entirely.

Try Subglow free →
ComponentRPC PollWebSocketRaw gRPCSubglow
Data delivery515ms95ms20ms20ms
Parse overhead28ms28ms28ms0ms
Bot logic5ms5ms5ms5ms
Total548ms128ms53ms25ms

Frequently asked questions

What is execution latency?

Execution latency is the total time from when a transaction confirms on Solana to when your bot can act on it. It includes data delivery time, parsing/deserialization time, and your bot's decision logic. Most discussions focus on wire latency, but parsing overhead (Borsh + Protobuf) often adds 15-30ms that people overlook.

Why does Borsh parsing take so long?

Solana instructions use Borsh binary serialization. When you receive raw gRPC data, each transaction arrives as a Protobuf message containing Borsh-encoded instruction data. Your bot must: decode Protobuf, identify the program, map instruction discriminators, deserialize Borsh fields, and resolve account indices. In JavaScript, this typically takes 15-30ms per transaction.

How does Subglow eliminate parsing overhead?

Subglow performs all Borsh deserialization, Protobuf decoding, instruction parsing, and account resolution server-side. Your bot receives clean JSON with human-readable fields like token, amount, buyer, and bonding_curve_pct — ready for your decision logic with zero client-side parsing.

Is this calculator accurate?

The calculator uses realistic defaults based on benchmarks across Solana infrastructure. Your actual latency depends on server location, code optimization, language runtime, and load. Adjust the sliders to match your setup for the most accurate comparison.

Stay ahead of the curve

Get Solana gRPC guides, infrastructure tips, and product updates delivered to your inbox. No spam — unsubscribe anytime.