---
title: "How Bullet uses ClickHouse Cloud to give DeFi's fastest exchange real-time analytics"
description: "Bullet operates a DeFi exchange for perpetuals, spot trading, and lending, with trades settled on a blockchain while users retain custody of their assets. Its sub-millisecond trading system produces more than 150 million event rows per hour, creating a need for historical indexing, real-time analytics, and operational monitoring beyond the blockchain’s state-machine model. ClickHouse Cloud consumes events from Kafka, stores them in dedicated tables, and supports user features, leaderboard and reward calculations, P&L charts, audit logs, and Grafana-backed alerts. Replacing Databricks reduced query latency from 10–15 seconds to milliseconds, improved data freshness from 1–2 hours to under 5 seconds, and removed a separate DynamoDB serving layer at comparable cost while handling 1,000 times more data across staging, testnet, and mainnet."
---

# How Bullet uses ClickHouse Cloud to give DeFi's fastest exchange real-time analytics

[Clickhouse](https://yomu.fyi/company/clickhouse) · Jul 15, 2026

**Type:** Problem & solution

## Summary

Bullet operates a DeFi exchange for perpetuals, spot trading, and lending, with trades settled on a blockchain while users retain custody of their assets. Its sub-millisecond trading system produces more than 150 million event rows per hour, creating a need for historical indexing, real-time analytics, and operational monitoring beyond the blockchain’s state-machine model. ClickHouse Cloud consumes events from Kafka, stores them in dedicated tables, and supports user features, leaderboard and reward calculations, P&L charts, audit logs, and Grafana-backed alerts. Replacing Databricks reduced query latency from 10–15 seconds to milliseconds, improved data freshness from 1–2 hours to under 5 seconds, and removed a separate DynamoDB serving layer at comparable cost while handling 1,000 times more data across staging, testnet, and mainnet.

## Context

Bullet needed fast, queryable history and live analytics for a blockchain exchange whose state-machine architecture does not retain historical events. Its previous Databricks-based setup produced 10–15-second queries and data that could be 1–2 hours stale, while the team also faced maintenance overhead from a multi-system data stack.

## Approach / What changed

A Rust indexer receives exchange events over WebSocket and batches them into Kafka on AWS MSK. Redpanda Connect classifies and routes events to dedicated Kafka topics, with a parallel S3 Glacier feed for disaster recovery. ClickHouse Cloud consumes the topics directly into dedicated tables, supporting user-facing analytics, monitoring snapshots, and refreshable materialized views.

## Takeaways

- ClickHouse Cloud’s columnar storage and compression suit Bullet’s append-only blockchain workload, which exceeds 150 million rows per hour and previously used uncompressed JSON in S3.
- Dedicated ClickHouse tables for event types allow each dataset to be sorted and ordered around its important fields, keeping queries fast while Kafka integration avoids custom ingestion wiring.
- Refreshable materialized views reduced leaderboard freshness from as much as 90 minutes in the previous setup to a 10-minute refresh interval, with the option to run every minute.

**Tags:** [AWS](https://yomu.fyi/topic/aws), [Kafka](https://yomu.fyi/topic/kafka), [Monitoring](https://yomu.fyi/topic/monitoring), [Performance](https://yomu.fyi/topic/performance), [Streaming](https://yomu.fyi/topic/streaming)

- Source: [Clickhouse](https://clickhouse.com/blog/bullet-real-time-analytics)
- Source URL: https://clickhouse.com/blog/bullet-real-time-analytics
- Ingested by Yomu: 2026-08-29T01:38:36.275Z

[Read original post](https://clickhouse.com/blog/bullet-real-time-analytics)