# Why Deploy on Evolving Edge?

> Why cloud customers deploy on Evolving Edge: how the network is architected, and what it delivers today.

Source: https://docs.evolvingedge.ai/why-deploy/

Evolving Edge lets you serve content and edge computing workloads from a network of thousands of independently-operated devices distributed across U.S. metro areas, instead of a handful of centralized data centers. The result is hyperlocal delivery: sub-10ms response times in served areas, compared to 70ms or more from traditional cloud CDN infrastructure. See [What is Evolving Edge?](https://docs.evolvingedge.ai/introduction/) for how the underlying network works.

## Who this is for

Businesses and developers who want low-latency edge computing capabilities in U.S. metro areas, without operating their own edge infrastructure.

## How the network is architected

Traditional cloud infrastructure centralizes computing in a small number of large data centers, so a request commonly travels several hundred to a few thousand miles round trip before a response comes back. Evolving Edge distributes that computing across three tiers instead:

| Tier | What it is | Where it sits | What it's for |
|---|---|---|---|
| Neighborhood nodes | Residential and small-business devices running the [node agent](https://docs.evolvingedge.ai/glossary/) | Inside the requesting user's MSA - often the same neighborhood | Primary delivery path; the source of the network's low-latency advantage |
| [Regional gateways](https://docs.evolvingedge.ai/glossary/) | Data center capacity Evolving Edge operates directly | Metro-to-regional scale | Overflow capacity and failover where neighborhood-node coverage is thin |
| AI Orchestration Layer | The control plane's routing and scoring system (see [Glossary](https://docs.evolvingedge.ai/glossary/)) | N/A - a coordination layer, not a physical tier | Routes each request to the best available node and predicts demand to pre-position content |

Because neighborhood-node density varies by MSA, actual latency in a given metro depends on how built-out the network is there - see [Coverage & Performance](https://docs.evolvingedge.ai/coverage-performance/) for what that looks like for a specific location. Regional gateways exist to keep the network functional everywhere else while neighborhood coverage grows.

## Capabilities

- **Hyperlocal delivery** - content served from inside the neighborhood or metro area (MSA) it's requested from, not routed across the country. See [MSA & POP](https://docs.evolvingedge.ai/glossary/) in the Glossary.
- **Distributed by design** - thousands of independently-operated nodes instead of a handful of hyperscale data centers, which means better redundancy against regional outages.
- **Transparent infrastructure** - the node software that serves your content is open source under the Mozilla Public License 2.0. You can review exactly how content is handled before trusting it to the network.
- **Growing capabilities** - the network currently focuses on CDN content delivery. **Edge AI Inference and Arbitrary Compute are on track to launch later this year**, letting you run inference and general-purpose workloads on the same distributed hardware that delivers your content today.

**Next:** [Getting Started >](https://docs.evolvingedge.ai/customer-getting-started/)
