This document describes the rollout strategies available in App Lifecycle Manager for
managing how releases are deployed across units in your SaaS offering.

When you manage distributed SaaS applications, deploying updates simultaneously
across all customer environments can lead to widespread outages if an unexpected
regression occurs. Rollout strategies protect your service availability by
controlling the pace, sequence, and spatial impact of software releases across
logical and physical failure domains. Configuring rollout strategies in
App Lifecycle Manager lets you automate gradual rollouts and introduce soak
times between deployment waves. You can also define error budgets that pause
problematic releases before they affect your customers.

## Overview of rollout strategies

App Lifecycle Manager provides configurable rollout strategies to manage how
units in your SaaS offering are updated. These rollout strategies follow
[Google Cloud's approach to change](https://docs.cloud.google.com/docs/cloud-approach-to-change.md.txt#safe-rollouts)
by applying consistent, staged deployment practices across your SaaS offering.

Use rollout strategies to minimize negative customer impact and isolate issues
to individual logical and physical failure domains. You define your rollout
strategy by creating a *rollout kind* that specifies one of the following
strategies:

- **One location at a time (simple)**
  (`Google.Cloud.Simple.OneLocationAtATime`): One location is updated at a time,
  with no wait between locations. Units are selected arbitrarily within a
  location, with a maximum of 5% of the location's units updated at a given
  time.

  This strategy is recommended for development environments and emergency
  scenarios.
- **All at once (simple)** (`Google.Cloud.Simple.AllAtOnce`): All locations
  begin updating at the same time. Within each location, units are selected
  arbitrarily, with a maximum of 5% of the location's units updated at a given
  time.

  This strategy is recommended for development environments and emergency
  scenarios.
- **Progressive (gradual)** (`Google.Cloud.Progressive.Gradual.v1`): Within each
  location, units are updated in percentage batches (1%, 10%, 20%, 30%, and
  \~40%) with soak times between batches. Across locations, the rollout
  progresses in exponential waves (for example, one location, then two, then
  four). Each wave is separated by a soak time, such as 18 hours. Units within a
  location are selected randomly.

  This strategy is designed for safe and predictable rollouts across multiple
  locations. It starts with a small footprint and gradually expands as
  confidence grows. This strategy is recommended for production environments
  with more than one location.
- **Progressive (single location)**
  (`Google.Cloud.Progressive.SingleLocation.v1`): Units are updated in static
  percentage batches (for example, 1%, 10%, 20%, 30%, and \~40%). Extended soak
  times between batches (for example, 18 hours) provide time for issue detection
  and limit the adverse impact of rollout changes.

  This strategy is tailored for SaaS offerings operating in a single location,
  prioritizing safety and caution. This strategy is recommended for production
  environments with a single location.

### Rollout strategy comparison and recommendations

The following table summarizes the key characteristics, soak time behaviors, and
recommended use cases for each rollout strategy:

| Rollout strategy | Value | Location behavior | Concurrency and batching | Soak time | Recommended use case |
|---|---|---|---|---|---|
| **One location at a time (simple)** | `Google.Cloud.Simple.OneLocationAtATime` | Single location sequentially | Up to 5% of units concurrently per location | None | Development environments and emergency rollouts |
| **All at once (simple)** | `Google.Cloud.Simple.AllAtOnce` | All locations concurrently | Up to 5% of units concurrently per location | None | Development environments, non-production testing, and emergency rollouts |
| **Progressive (gradual)** | `Google.Cloud.Progressive.Gradual.v1` | Multi-location in exponential waves (1, 2, 4 locations) | Percentage batches (1%, 10%, 20%, 30%, and \~40%) | Soak times between batches; 18 hours between location waves | Production environments with multiple locations |
| **Progressive (single location)** | `Google.Cloud.Progressive.SingleLocation.v1` | Single location | Percentage batches (1%, 10%, 20%, 30%, and \~40%) | Extended soak times (18 hours) between batches | Production environments with a single location |

## Configure rollout kinds and safeguards

When you [create a rollout kind](https://docs.cloud.google.com/app-lifecycle-manager/update-saas-offering.md.txt#create_a_rollout_kind),
you associate a rollout strategy with a target `UnitKind` and can configure
additional safeguards:

- **Error budget (`error_budget`)** : Automatically pauses a rollout when failed
  unit updates in a location exceed your configured threshold (`allowed_count`
  or `allowed_percentage`). If you don't configure an error budget,
  App Lifecycle Manager attempts to update all units regardless of the number
  of failures.

- **Unit filter (`unit_filter`)**: Applies a Common Expression Language (CEL)
  filter string to restrict the eligible population of units targeted by the
  rollout.

- **Regional scope** : Supported rollout locations are determined by the
  available regions configured on your SaaS offering. For details on how regions
  affect rollout propagation, see
  [App Lifecycle Manager regionalization](https://docs.cloud.google.com/app-lifecycle-manager/regionalization.md.txt).

## What's next

- Learn how to create releases, rollout kinds, and rollouts in
  [Update a SaaS offering](https://docs.cloud.google.com/app-lifecycle-manager/update-saas-offering.md.txt).

- Understand how regions and resource replication work in
  [App Lifecycle Manager regionalization](https://docs.cloud.google.com/app-lifecycle-manager/regionalization.md.txt).

- Learn more about App Lifecycle Manager in the
  [App Lifecycle Manager overview](https://docs.cloud.google.com/app-lifecycle-manager/overview.md.txt).