Today, we're excited to announce a new research project we've been kicking around at Kong: Kong Embedded!

If you've used the Kong Gateway before or heard us talk about it, one of the things we're very proud of is that Kong Gateway uses a very small resource footprint. It's a small download in size, is blazingly fast on even constrained hardware, and uses very little memory. On traditional enterprise hardware, this means you can get more requests per second through the gateway on the same amount of hardware, which helps to lower the total cost of ownership of the Kong Gateway.

Having this low resource usage requirement has another advantage that we're now exploring: the ability to run directly on low-power consumer and edge-deployed devices like those in industrial IoT and enterprise edge device settings. And that's exactly what Kong Embedded brings to the table. Kong Embedded is a new way of deploying Kong Enterprise on edge devices. It's lightweight and can be compiled for mobile architectures like ARM CPUs. It can run without a database to further reduce the resource requirements and simplify the software architecture.

Kong Embedded fits into architectures you may have never thought possible with API gateways and opens up entirely new possibilities:

• - Secure devices and APIs that can live directly on the edge through one of Kong Enterprise's many authentication and authorization plugins
• - Enable consistent, secure, and observable device-to-device communications through APIs
• - Rate-limit access to device APIs through the gateway to preserve precious CPU
• - Implement custom logic for when specific APIs are called on edge devices

We leave it to you to come up with even more exciting use cases than we may be thinking of and playing around with! Kong Embedded is still in the early stages of research and development and is currently only available on an invite-only basis. If you'd like to sign up to hear more and ask to be included in product evaluation, head on over to our [Kong Embedded page](https://konghq.com/kong-embedded)Kong Embedded page to let us know a bit more detail about your use case, and we'll get back to you!