Skip to main content
velcio

How your code becomes
a running application.

Connect GitHub or GitLab, choose a deployment branch and configure your app. Push to that branch to build a new version. Velcio sends it traffic after its health checks pass.

Deploy your appPricing
  1. Connect

    Choose a repository and branch

    Connect your source control, then add environment values and secrets.

  2. Build

    Build a versioned image

    Velcio uses your Dockerfile or detects a compatible buildpack.

  3. Place

    Start regional microVMs

    Choose your regions and set the baseline replica count for each one.

  4. Operate

    Match capacity to demand

    Autoscaling adds replicas when demand rises. Velcio replaces unhealthy instances.

A baseline that stays in place.

The replica counts you set in each region form your minimum capacity. Autoscaling adds temporary replicas above that baseline as demand changes, up to 10 replicas total on self-serve plans.

Region A

2 baseline replicas

Region B

1 baseline replica

As demand rises

Temporary replicas in configured regions
Up to 10 replicas total, including the baseline

Illustrative configuration. Baseline counts stay in place. New replicas receive traffic only after passing health checks.
Routing details

Requests from the same session may reach different instances, so store shared sessions outside the app.

  1. Request

    A request arrives at your app’s public HTTPS address.

  2. Region

    Velcio prefers the nearest region with healthy replicas.

  3. Origin

    Traffic is distributed across hosts according to their ready replica counts.

  4. Replica

    The host forwards the request to a ready app instance.

  5. Retry

    If an eligible request with no body fails early, Velcio can retry it on one other host.

Autoscaling adds capacity only in configured regions and never reduces their baseline counts. During recovery, eligible Standard and Pro replicas may run temporarily in the recovery region selected for that replica.

A failed instance gets replaced.

Velcio checks the application's health endpoint, removes unhealthy replicas from routing and restores the configured count.

  1. Detect

    A health check fails.

  2. Isolate

    The unhealthy instance leaves routing.

  3. Replace

    A fresh instance starts from the application image.

  4. Readmit

    Traffic resumes after health checks pass.

Other healthy replicas keep serving. Local files and memory are not recovered.

All configured regions are active. If one has no healthy origin, Velcio uses another configured healthy region.

Eligible apps can temporarily run from a prepared copy elsewhere, then return when the original region is healthy.

Extra protection when a region goes down.

Each eligible Standard and Pro replica can use one selected recovery region during an outage, then return when its original region recovers. Cross-region recovery can also be turned off per replica.

Standard uses spare capacity. Pro reserves capacity once protection is ready. Recovery never falls back outside the selected source and recovery region pair. Normal hosting charges continue.

Same public application URL
  1. Normal
    Source region

    Serving traffic

    Selected recovery region

    Prepared copy, stopped

  2. Outage
    Source region

    Unavailable

    Selected recovery region

    Starts, passes health checks, serves traffic

  3. Return
    Source region

    Stable and healthy; traffic returns

    Selected recovery region

    Temporary runtime stops

For an eligible replica with recovery enabled. Standard uses spare capacity; Pro reserves capacity once protection is ready. Recovery stays within the selected region pair.
Recovery details

Regional recovery is available for eligible stateless web apps. After a sustained regional failure, Velcio starts a prepared copy in that replica's selected recovery region with the current configuration and secrets. Traffic moves only after the app passes its health check.

Traffic returns when the original region is stable and healthy. Stopping the app also pauses recovery.

Standard uses spare capacity after Pro; a prepared copy does not guarantee capacity. Pro gets priority and reserves capacity once protection is ready.

Recovery runs only in the selected recovery region and can be disabled per replica. Databases, storage and allowlists must work from that region. Local files, memory and open connections are not recovered.

In a simulated regional blackout, recovery completed in approximately two minutes. This is a test result, not an uptime or recovery-time guarantee.

Eligible paid production Web Services have a 99.9% monthly availability commitment from 1 December 2026. It is measured separately from recovery timing on the platform hostname and verified active custom hostnames using a two-of-three external-region quorum. Read the Cloud SLA.

What your app needs to provide

Your application
Repository, deploy branch, runtime values, a process bound to PORT, and a health endpoint.
What Velcio manages
Builds, isolated app instances, runtime secrets, HTTPS and routing. All plans replace unhealthy instances; regional recovery depends on your plan and app eligibility.

Persistent disks are not supported. Store durable data and shared sessions outside the app. Replacement instances start fresh.

Watch the console tour Create, deploy, schedule, add resources

Deploy your first application.

Connect GitHub or GitLab, choose a branch and set up your deployment.

Deploy your appCompare plans