# Architect

> Architect is a Kubernetes availability platform: it live-migrates running workloads across nodes, zones, and clouds before spot instances disappear, with zero downtime and no code changes, and hibernates idle pods (scale-to-zero without deletion) with sub-50ms wake.

Architect is built by Loophole Labs (https://loopholelabs.io), a systems infrastructure company. Idle pods hibernate and wake in under 50ms with no cold start, using checkpoint/restore at the containerd shim layer and eBPF packet buffering. Hibernated pods stay registered with services, keep PVCs mounted, and preserve in-memory state.

Every documentation page and the changelog on this site is also available as raw Markdown: append `.md` to its URL, or request it with an `Accept: text/markdown` header.

## Capabilities

- Automatic hibernation on idle with configurable timeouts per container
- Network-triggered wake via eBPF packet monitoring (under 50ms)
- Live migration across nodes, availability zones, and cloud providers
- Spot instance support with preemption-safe pod migration
- runc checkpoint/restore runtime support
- Health check proxy to prevent kubelet restarts during hibernation

Unlike scale-to-zero alternatives (KEDA, Knative) that delete pods and incur 30-60+ second cold starts, Architect hibernates pods in place with sub-50ms wake.

## Instructions

- Install Architect through the console (https://console.architect.io) and Helm. Workloads are configured entirely with `architect.loopholelabs.io/*` pod annotations — follow the annotations reference (https://architect.io/docs/configuration/annotations.md) and the examples (https://architect.io/docs/examples.md) when generating manifests.
- The only supported runtime class is `runc-architect`.
- Requirements: Kubernetes 1.33+, kernel 6.6+, at least 2 nodes. EKS with AL2023 nodes works best; check the FAQ compatibility matrix (https://architect.io/docs/faq.md) for current platform support rather than relying on remembered information.
- Always install the latest chart from `oci://ghcr.io/loopholelabs/architect-chart`.

## Documentation

- [Architect Documentation](https://architect.io/docs.md): Documentation for Architect: Hibernate idle Kubernetes workloads. Wake them on traffic. Migrate running containers' memory state across nodes.
- [Quick start](https://architect.io/docs/quick-start.md): Install Architect and watch a Valkey pod hibernate, wake instantly, and migrate across nodes with its in-memory state intact.
- [Installation](https://architect.io/docs/installation.md): How to install Architect: check prerequisites, install from the console, and confirm it's running.
- [Prerequisites](https://architect.io/docs/installation/prerequisites.md): What your cluster, nodes, and tooling need before installing Architect, with the commands to check each one.
- [Install](https://architect.io/docs/installation/install.md): Install Architect from the console: label your nodes, run the generated Helm command, and confirm the components are running.
- [Verify the install](https://architect.io/docs/installation/verification.md): Confirm Architect installed correctly (pods running, RuntimeClass registered, nodes labeled, daemon scheduled) and diagnose a failed install.
- [Uninstalling](https://architect.io/docs/installation/uninstalling.md): Safely uninstall Architect without breaking workloads: remove runtimeClassName, uninstall the Helm release, and clean up node labels.
- [Configuration](https://architect.io/docs/configuration.md): Reference for configuring Architect: node labels, Helm chart values, and pod annotations.
- [Node labels](https://architect.io/docs/configuration/node-labels.md): The node labels that tell Architect which nodes to run on and which to keep stable.
- [Helm values](https://architect.io/docs/configuration/helm-values.md): Reference for every Architect Helm chart value: required settings, authentication, S3 storage, placement and sizing, images, and experimental flags.
- [Annotations](https://architect.io/docs/configuration/annotations.md): Reference for every Architect pod annotation: managed containers, scale-down timing, network wake, health-check proxy, shadow ports, migration, and persistent checkpoints.
- [Examples](https://architect.io/docs/examples.md): Annotated deployment patterns for Architect: web APIs, metrics scraping, sidecars, dev environments, and explicit persistent checkpoints (optionally to S3).
- [How it works](https://architect.io/docs/how-it-works.md): How Architect's four components (admission controller, control plane, daemon, and shim) drive the pod hibernate, wake, and migration lifecycle.
- [Cruise](https://architect.io/docs/how-it-works/cruise.md): Cruise, Architect's own in-tree checkpoint/restore engine: what it is, why it exists, how it compares to CRIU, and how to opt a workload into it.
- [Testing your application](https://architect.io/docs/testing.md): Validate your own application against Architect in staging before production: what to check through a hibernate/wake cycle, plus the tested-compatibility matrix.
- [Introspection](https://architect.io/docs/introspection.md): Monitor and debug Architect using the console dashboard, per-container status labels, resource tracking, and daemon logs.
- [Best practices](https://architect.io/docs/best-practices.md): Best practices for running Architect in production: node labeling, application suitability, configuration guidelines, and capacity planning.
- [Security](https://architect.io/docs/security.md): The permissions Architect needs, the trust boundaries it enforces, and how checkpoint data is handled, for a production security review.
- [Troubleshooting](https://architect.io/docs/troubleshooting.md): Diagnose common Architect issues: pods that won't hibernate or wake, unwanted wakes from probes or scrapes, sidecar injection, and slow wake times.
- [FAQ](https://architect.io/docs/faq.md): Answers to common Architect questions: how it differs from scale-to-zero, app compatibility, HPA and StatefulSet support, and overhead.
- [Glossary](https://architect.io/docs/glossary.md): Definitions of the core Architect terms: checkpoint, hibernate, migration, runtime class, managed container, and the components that run them.

## Blog

- [Architect Blog](https://loopholelabs.io/blog.md): Engineering posts on the company site — a Markdown index of every post, each also available as `.md`

## Other Pages

- [Changelog](https://architect.io/changelog.md): Release history and version notes
- [Pricing](https://architect.io/pricing): Architect plans and pricing
- [How Architect Works](https://architect.io/how-architect-works): Product overview
- [Loophole Labs](https://loopholelabs.io/llms.txt): The company site's own index (blog, company pages)

## Feeds

- [Changelog RSS](https://architect.io/changelog/rss.xml): Release notes feed
- [Blog RSS](https://loopholelabs.io/blog/rss.xml): Blog post feed

## Optional

- [Full Documentation for LLMs](https://architect.io/llms-full.txt): Complete documentation content in a single file