> ## Documentation Index
> Fetch the complete documentation index at: https://docs.duohub.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Improve Latency

> Tips on how to get the best latency from your application

## Minimize Client-Server Communication

Avoid making multiple calls between the client (user's machine) and server (cloud processes). Instead, aim to:

* Make a single call to the cloud
* Keep cross-service communication within the cloud infrastructure
* Batch requests where possible

## Consider Geographic Proximity

Location matters significantly for latency:

* duohub automatically routes queries to the nearest server
* Ensure your servers are strategically located close to your user base
* Use geo-distributed infrastructure when possible

## Implement Data Streaming

Leverage streaming capabilities:

* Stream data whenever possible
* Begin processing data before the entire stream completes
* Enables parallel processing and faster response times

## Utilize Caching Effectively

Implement caching strategies:

* Use caching for frequently accessed data
* Consider Redis cache for high-performance caching needs
* Implement appropriate cache invalidation policies

## Choose Optimized Services

Select services designed for low latency:

* Use platforms like duohub, Vercel, AWS Lambda @ Edge, etc. that are optimized for performance
* Look for services that offer:
  * Geo-distribution
  * Built-in caching
  * Optimized network routing