> For the complete documentation index, see [llms.txt](https://docs.algenta.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.algenta.ai/deploy-and-operate/deployments-and-nodes.md).

# Deployments & nodes

Algenta can run your org on the shared managed pool or on an **isolated deployment** in your own cloud region. This page covers the deployment lifecycle — list regions, request a deployment, check status, read cost, and deprovision — and how self-hosted **worker nodes** register and heartbeat so they appear in your infrastructure view.

{% hint style="info" %}
Requesting or deprovisioning a deployment requires the **owner** role; reading cost requires **admin**. Listing regions, reading the current deployment, and node registration read the authenticated org. Examples send `Authorization: Bearer $ALGENTA_API_KEY`; swap `https://api.algenta.ai` for `http://localhost:8000` on a self-hosted engine.
{% endhint %}

## Provision and manage a deployment

{% stepper %}
{% step %}

### List providers and regions

Every deployment targets a provider and a region. Fetch the current options first — the `id` fields are what you pass when creating a deployment.

```bash
curl -sS "https://api.algenta.ai/v1/deployments/regions" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

The four providers are `algenta_shared` (the default multi-tenant pool, no setup), `aws`, `azure`, and `gcp`, each with a list of regions:

```json
{
  "providers": [
    { "id": "algenta_shared", "name": "Algenta Managed", "regions": [ ... ] },
    { "id": "aws", "name": "Amazon Web Services", "regions": [ { "id": "us-east-1", "label": "US East (N. Virginia)", "flag": "🇺🇸" } ] },
    { "id": "azure", "name": "Microsoft Azure", "regions": [ ... ] },
    { "id": "gcp", "name": "Google Cloud Platform", "regions": [ ... ] }
  ]
}
```

{% endstep %}

{% step %}

### Request a deployment

Create the deployment with a `provider` and `region`. Optional `config` keys (`instance_size`, `storage_gb`, `max_concurrent_runs`, `auto_scale_max`) size it. The call returns `202 Accepted` immediately — provisioning is asynchronous and typically takes a few minutes.

```bash
curl -sS -X POST "https://api.algenta.ai/v1/deployments" \
  -H "Authorization: Bearer $ALGENTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "provider": "aws",
    "region": "us-east-1",
    "config": { "instance_size": "medium", "storage_gb": 100 }
  }'
```

An unknown provider or region returns `422` (`invalid_provider` / `invalid_region`), and an org that already has an active or in-progress deployment returns `409` `deployment_exists` — deprovision the existing one first.
{% endstep %}

{% step %}

### Poll status

Read the org's current deployment. It returns `null` while you are on the shared pool, otherwise the deployment record with its `status`.

```bash
curl -sS "https://api.algenta.ai/v1/deployments" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

```json
{
  "deployment_id": "d1e2f3a4-...",
  "org_id": "0123abcd-...",
  "provider": "aws",
  "region": "us-east-1",
  "status": "provisioning",
  "endpoint_url": null,
  "cost_usd_month": 0.0,
  "billable_cost_usd_month": 0.0,
  "billing_markup_pct": 20.0,
  "created_at": "2026-07-05T09:12:44Z",
  "provisioned_at": null,
  "error_message": null
}
```

`status` moves `requested` → `provisioning` → `active`. When it is `active`, `endpoint_url` is populated and your API calls route automatically to the isolated deployment. A failed provision reports `status: "failed"` with an `error_message`.
{% endstep %}

{% step %}

### Read monthly cost

Check the current-month cloud cost for a deployment. `cost_usd_month` is the underlying cloud cost and `billable_cost_usd_month` applies the pass-through `billing_markup_pct`.

```bash
curl -sS "https://api.algenta.ai/v1/deployments/$DEPLOYMENT_ID/cost" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

The response echoes `provider`, `region`, the current `year` and `month`, both cost figures, the markup percentage, and `last_updated`.
{% endstep %}

{% step %}

### Deprovision

Tear the deployment down and revert the org to the shared pool. Like create, this returns `202` and runs asynchronously.

```bash
curl -sS -X DELETE "https://api.algenta.ai/v1/deployments/$DEPLOYMENT_ID" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

It responds `{ "status": "deprovisioning", "deployment_id": "..." }`. An already-torn-down deployment returns `409` `already_deprovisioned`; an unknown id returns `404` `not_found`.
{% endstep %}
{% endstepper %}

{% hint style="success" %}
**Expected result** — `POST /v1/deployments` returns `202`, `GET /v1/deployments` shows the deployment advancing to `status: "active"` with a populated `endpoint_url`, and `GET /v1/deployments/{deployment_id}/cost` returns the current month's `cost_usd_month` and `billable_cost_usd_month`.
{% endhint %}

## Register self-hosted worker nodes

A self-hosted worker registers with the engine on startup and then heartbeats, so it shows up alongside the primary node in `GET /v1/infrastructure`.

{% stepper %}
{% step %}

### Register the node

Call register on startup. All fields are optional — omit `node_id` and one is generated for you.

```bash
curl -sS -X POST "https://api.algenta.ai/v1/nodes/register" \
  -H "Authorization: Bearer $ALGENTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "node_id": "worker-eu-1",
    "region": "eu-west-1",
    "engine_version": "0.3.1",
    "cpu_count": 8,
    "memory_gb": 32,
    "capabilities": ["large-memory"]
  }'
```

The `201` response returns the `node_id` and confirms where to send heartbeats.
{% endstep %}

{% step %}

### Send heartbeats

Send a heartbeat about every 30 seconds to keep the node live and report its current load. `status` may be `live`, `idle`, `error`, or `updating`.

```bash
curl -sS -X POST "https://api.algenta.ai/v1/nodes/heartbeat" \
  -H "Authorization: Bearer $ALGENTA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "node_id": "worker-eu-1",
    "cpu_pct": 24.1,
    "memory_gb": 3.5,
    "jobs_per_sec": 251.0,
    "status": "live"
  }'
```

A heartbeat for an unregistered node returns `404` `node_not_found`; a node owned by a different org returns `403`.
{% endstep %}

{% step %}

### View infrastructure and deregister

See the primary node plus your registered workers, with live engine health and latency.

```bash
curl -sS "https://api.algenta.ai/v1/infrastructure" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

Each node reports `node_id`, `status`, `engine_version`, `cpu_pct`, `memory_gb`, `jobs_per_sec`, `region`, and `last_heartbeat`. On graceful shutdown, deregister the node:

```bash
curl -sS -X DELETE "https://api.algenta.ai/v1/nodes/worker-eu-1" \
  -H "Authorization: Bearer $ALGENTA_API_KEY"
```

{% endstep %}
{% endstepper %}

{% hint style="success" %}
**Expected result** — `POST /v1/nodes/register` returns `201`, subsequent heartbeats return `{ "node_id": "...", "accepted": true }`, and the node appears in `GET /v1/infrastructure` with a recent `last_heartbeat` and its reported metrics.
{% endhint %}

## Next steps

<table data-view="cards"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Monitor the engine</strong></td><td>Health checks, metrics, and latency for your nodes.</td><td><a href="/pages/70qQAu9mBa5YatEkZbKV">/pages/70qQAu9mBa5YatEkZbKV</a></td></tr><tr><td><strong>Self-host the engine</strong></td><td>Run the runtime and workers on your own infrastructure.</td><td><a href="/pages/tWtESdNlxR5ZqT7F3czC">/pages/tWtESdNlxR5ZqT7F3czC</a></td></tr><tr><td><strong>Track plans and usage</strong></td><td>How deployment cost and markup roll into billing.</td><td><a href="/pages/LtqjHw8bwZ1rd0qQ3xgl">/pages/LtqjHw8bwZ1rd0qQ3xgl</a></td></tr></tbody></table>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.algenta.ai/deploy-and-operate/deployments-and-nodes.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.