🚀 Deribit’s New Matching Engine: From Milliseconds to Microseconds
Deribit’s new matching engine is built to process high order volumes with faster and more predictable response times.
Measured results following the rollout:
➡️ Median request-response latency fell from 4.7 milliseconds to 76 microseconds, an improvement of roughly 61×.
➡️ P99 latency, covering the slowest 1% of responses, fell from 124 milliseconds to 234 microseconds, an improvement of roughly 531×.
Read the full article and peer-reviewed technical paper here: