# ElastiCache Serverless for Valkey now has public endpoints, no VPN or bastion needed!

Hi everyone, it's Shii-chan! Ever wanted to connect to your cache straight from your own machine? Today's news makes that possible!

## What was announced?

According to AWS What's New, Amazon ElastiCache Serverless for Valkey now supports public endpoints. You can connect to your cache directly from a laptop, a serverless function, or any application outside AWS, without a VPN, a bastion host, or an SSH tunnel.

## The story so far

Until now, connecting was built around being inside a VPC, so reaching a cache from your own machine or an outside environment meant setting up a VPN, bastion, or tunnel. With a public endpoint there is no VPC to configure and no infrastructure to provision, and you can create one in under a minute.

## What changes

The announcement highlights these uses:

- Rapid prototyping
- Connecting AI coding tools and agents straight to your cache
- Adding caching to workloads that can't reach a VPC

## Dive Deep

Here are the security and connectivity details:

- Every connection uses IAM authentication over TLS 1.3, so there is no password to store or rotate
- Valkey GLIDE 2.2 or later has built-in IAM support
- Other Valkey clients can be paired with the Developer Toolkit for ElastiCache, an open-source library that generates and refreshes IAM authentication tokens
- Requires a Valkey 9.0 or later serverless cache
- Create one from the AWS Management Console, AWS SDK, or AWS CLI
- Available in all commercial AWS Regions and the China Regions
- No additional charge beyond standard ElastiCache Serverless pricing

## Wrap-up

- ElastiCache Serverless for Valkey supports public endpoints
- Connect from outside AWS without a VPN, bastion, or SSH tunnel
- IAM authentication over TLS 1.3 means no passwords to manage
- Requires Valkey 9.0 or later, and a client such as Valkey GLIDE 2.2 or later
- No extra charge

This is for developers who want to try a cache quickly from their own machine, and for anyone who wants AI agents to use a cache!