# We benchmarked 100,000 database calls and caching changed everything

Canonical: https://brew.new/browse/templates/email/pt1_k97zjpdw99mvhft6yvxrka81fd8e2dxa

Brand: appwrite.io
Category: newsletter

![Preview of We benchmarked 100,000 database calls and caching changed everything](https://cdn.brew.new/email-preview-2c442ff111b9ac73-tracking_r57g8vh94brtjz3h67cvv3ketx8dt6t0-1788996861327.png)

## Email content

Hello Appwriters,

Most applications repeatedly execute the same queries. Dashboards, feeds, and product listings often fetch data that does not change between requests, yet the database still processes each one.

We wanted to measure the impact of fixing that.

So we ran a 100,000-request benchmark on Appwrite Databases, comparing performance with and without TTL caching.

The results were significant. Average latency dropped by more than 50%, throughput nearly doubled, and repeated queries became dramatically cheaper to serve.

TTL caching works by storing the result of a query after the first request and serving identical requests from memory within a defined time window. Because responses are served directly from memory, they are significantly faster than regular database reads, while also reducing load on the database.

In the full breakdown, we go deeper into how caching affects tail latency, where it provides the most benefit, and when it is worth enabling in real workloads.

Read the full benchmark:

https://appwrite.io/blog/post/ttl-cache-performance-benchmark

Happy coding!

Eldad Fux

Founder & CEO

@eldadfux

You're receiving this message as a subscriber of Appwrite. If you prefer to not receive these messages anymore, feel free to unsubscribe. Our postal address is 251 Little Falls Drive, Wilmington, Delaware, 19808, United States.

[Open and remix this design](https://brew.new/browse/templates/email/pt1_k97zjpdw99mvhft6yvxrka81fd8e2dxa)

[Browse email designs](https://brew.new/browse/templates)