---
title: "How Jua delivers the world’s most accurate physics simulations 3x faster with ClickHouse Cloud"
description: "Jua uses ClickHouse Cloud to deliver data from EPT-2, its physics foundation model, to energy traders and grid operators faster than its previous file-based pipeline. Jua’s forecasts combine outputs from multiple models and add roughly four terabytes of new data daily, producing about one petabyte compressed and over 10–12 petabytes uncompressed. Replacing Zarr and other file-based storage plus manual pre-fetching with ClickHouse Cloud reduced flagship forecast delivery from one hour to 20 minutes, with about 20 seconds of upload overhead. The platform also cut compute costs by roughly a third, reduced two-year single-location historical queries from hours to seconds, and supports both point lookups and regional aggregations; Jua selected the managed service for operational simplicity, separated storage and compute, and cost efficiency."
---

# How Jua delivers the world’s most accurate physics simulations 3x faster with ClickHouse Cloud

[Clickhouse](https://yomu.fyi/company/clickhouse) · Jun 30, 2026

**Type:** Problem & solution

## Summary

Jua uses ClickHouse Cloud to deliver data from EPT-2, its physics foundation model, to energy traders and grid operators faster than its previous file-based pipeline. Jua’s forecasts combine outputs from multiple models and add roughly four terabytes of new data daily, producing about one petabyte compressed and over 10–12 petabytes uncompressed. Replacing Zarr and other file-based storage plus manual pre-fetching with ClickHouse Cloud reduced flagship forecast delivery from one hour to 20 minutes, with about 20 seconds of upload overhead. The platform also cut compute costs by roughly a third, reduced two-year single-location historical queries from hours to seconds, and supports both point lookups and regional aggregations; Jua selected the managed service for operational simplicity, separated storage and compute, and cost efficiency.

## Context

Jua needed to make large physics-simulation and weather-forecast datasets available to customers faster. Its previous file-based pipeline required pre-fetching, caching, and manual notification, creating an access delay even when forecasts were generated quickly. A small engineering team also found self-hosting and maintaining the storage infrastructure too burdensome.

## Approach / What changed

Jua replaced its file-based and Zarr-based forecast pipeline with ClickHouse Cloud. The service stores the compressed simulation data, handles uploads and caching, and supports both point queries for specific locations and broader regional aggregations or analytics. Jua chose the managed deployment partly because its separation of storage and compute reduced replication and maintenance requirements.

## Takeaways

- ClickHouse Cloud reduced Jua’s flagship EPT-2 forecast delivery time from one hour to 20 minutes, while upload overhead was about 20 seconds.
- Jua’s data volume is about four terabytes of new data per day, totaling roughly one petabyte compressed and over 10–12 petabytes uncompressed.
- Historical queries that previously required hours of manual file searches can now return two years of data for a single location in seconds, and the same system supports regional aggregations.

**Tags:** [Data Pipelines](https://yomu.fyi/topic/data-pipelines), [Machine Learning](https://yomu.fyi/topic/machine-learning), [Performance](https://yomu.fyi/topic/performance), [Scalability](https://yomu.fyi/topic/scalability)

- Source: [Clickhouse](https://clickhouse.com/blog/jua-physics-foundation-model)
- Source URL: https://clickhouse.com/blog/jua-physics-foundation-model
- Ingested by Yomu: 2026-08-29T01:37:12.325Z

[Read original post](https://clickhouse.com/blog/jua-physics-foundation-model)