> For the complete documentation index, see [llms.txt](https://developer.harness.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.harness.io/resilience-testing/chaos-engineering/new-to-chaos-engineering/on-premise-vs-saas.md).

# On-premise vs SaaS

Harness Chaos Engineering offers two deployment options to meet your organization's requirements. Choose between the fully managed SaaS solution or deploy the Self-Managed Enterprise Edition on-premises.

## Harness Chaos Engineering SaaS <a href="#harness-chaos-engineering-saas" id="harness-chaos-engineering-saas"></a>

### Overview <a href="#overview" id="overview"></a>

Harness SaaS is a fully managed, cloud-hosted solution that provides immediate access to chaos engineering capabilities without infrastructure setup or maintenance overhead.

### Key Features <a href="#key-features" id="key-features"></a>

* **Quick Setup**: Start running chaos experiments within minutes
* **Managed Infrastructure**: No need to manage underlying platform components
* **Automatic Updates**: Always access the latest features and patches
* **Enterprise ChaosHub**: Built-in access to 200+ pre-built chaos faults
* **AI Support**: AI Development Assistant for creating governance policies

### What's Included <a href="#whats-included" id="whats-included"></a>

* **Complete Platform**: Full chaos engineering capabilities out-of-the-box
* **Enterprise ChaosHub**: Access to comprehensive fault library
* **Harness AI**: AI assistant for policy creation and optimization
* **Managed Services**: Platform maintenance handled by Harness
* **Standard Integrations**: Built-in connectors for popular tools

### Ideal For <a href="#ideal-for" id="ideal-for"></a>

* Organizations wanting to get started quickly
* Teams focused on application development over infrastructure
* Companies with standard compliance requirements
* Development and testing environments

## Self-Managed Enterprise Edition (On-Premise) <a href="#self-managed-enterprise-edition-on-premise" id="self-managed-enterprise-edition-on-premise"></a>

### Overview <a href="#overview" id="overview"></a>

The Self-Managed Enterprise Edition allows you to run Harness Chaos Engineering entirely within your own infrastructure, providing complete control over your chaos engineering environment.

### Key Features <a href="#key-features" id="key-features"></a>

* **Complete Data Control**: All data remains within your infrastructure
* **Feature Parity**: Same capabilities as SaaS with minor timeline differences
* **Network Isolation**: Deploy in air-gapped or restricted environments
* **Custom Configuration**: Tailor the platform to your specific needs
* **Enterprise Integration**: LDAP, SSO, and custom authentication

### Architecture Components <a href="#architecture-components" id="architecture-components"></a>

* **Chaos Control Plane**: Manages experiments and orchestration
* **Database Layer**: MongoDB and TimescaleDB for data storage
* **Monitoring Stack**: Prometheus and Grafana for observability
* **Authentication Service**: User management and access control
* **Load Balancer**: NGINX or Istio virtual services

### Infrastructure Requirements <a href="#infrastructure-requirements" id="infrastructure-requirements"></a>

#### Minimum Requirements <a href="#minimum-requirements" id="minimum-requirements"></a>

* **CPU**: 16 cores total
* **Memory**: 32 GB RAM
* **Storage**: 500 GB SSD
* **Kubernetes**: v1.21+ (any CNCF-certified distribution)

#### Recommended for Production <a href="#recommended-for-production" id="recommended-for-production"></a>

* **CPU**: 32+ cores
* **Memory**: 64+ GB RAM
* **Storage**: 1+ TB NVMe SSD
* **High Availability**: Multi-node setup with redundancy

### Deployment Options <a href="#deployment-options" id="deployment-options"></a>

* **Cloud**: Deploy on public cloud Kubernetes services
* **Connected**: On-premises with internet connectivity
* **Air-gapped**: Completely isolated environments
* **Signed Certificates**: Custom certificate management

### Limitations Compared to SaaS <a href="#limitations-compared-to-saas" id="limitations-compared-to-saas"></a>

* **Enterprise ChaosHub**: Not connected by default
  * Contact [Harness Support](mailto:support@harness.io) for access
  * Can be added as a [custom ChaosHub](/resilience-testing/chaos-engineering/use-chaos-engineering/chaoshub.md)
* **Harness AI**: AI Development Assistant not available
* **Updates**: Manual deployment of platform updates required
* **Maintenance**: Customer responsibility for platform operations

## Feature Comparison <a href="#feature-comparison" id="feature-comparison"></a>

| Feature                   | SaaS                  | Self-Managed Enterprise |
| ------------------------- | --------------------- | ----------------------- |
| **Setup Time**            | Minutes               | Days to Weeks           |
| **Platform Maintenance**  | Managed by Harness    | Customer Responsibility |
| **Feature Updates**       | Automatic             | Manual Deployment       |
| **Enterprise ChaosHub**   | ✅ Included            | ❌ Requires Setup        |
| **Harness AI**            | ✅ Available           | ❌ Not Available         |
| **Data Location**         | Harness Cloud         | Customer Infrastructure |
| **Air-gapped Deployment** | ❌ Not Supported       | ✅ Supported             |
| **Custom Integrations**   | Standard APIs         | Full API Access         |
| **Compliance Control**    | Shared Responsibility | Full Customer Control   |

## Supported Platforms (Both Deployments) <a href="#supported-platforms-both-deployments" id="supported-platforms-both-deployments"></a>

Both SaaS and Self-Managed Enterprise Edition support the same target platforms:

### Kubernetes Distributions <a href="#kubernetes-distributions" id="kubernetes-distributions"></a>

* Amazon EKS (k8s 1.21+)
* Azure AKS (k8s 1.21+)
* Google GKE (k8s 1.21+)
* Red Hat OpenShift (k8s 1.21+)
* Rancher (k8s 1.21+)
* VMware Tanzu (k8s 1.21+)
* Self-managed clusters (k8s 1.21+)

### Operating Systems <a href="#operating-systems" id="operating-systems"></a>

* Ubuntu 16.04+
* Red Hat Enterprise Linux 7+
* CentOS 7+
* Debian 10+
* Fedora 30+
* openSUSE Leap 15.4+

### Chaos Fault Categories <a href="#chaos-fault-categories" id="chaos-fault-categories"></a>

Both deployments support the same fault types across Linux, Kubernetes, Windows, VMware, AWS, GCP, and Azure platforms.

## Making the Decision <a href="#making-the-decision" id="making-the-decision"></a>

### Choose SaaS If You: <a href="#choose-saas-if-you" id="choose-saas-if-you"></a>

* Want to get started quickly with minimal setup
* Prefer managed services and automatic updates
* Need immediate access to Enterprise ChaosHub
* Want to use Harness AI for policy creation
* Have standard security and compliance requirements

### Choose Self-Managed Enterprise Edition If You: <a href="#choose-self-managed-enterprise-edition-if-you" id="choose-self-managed-enterprise-edition-if-you"></a>

* Have strict data sovereignty requirements
* Need to deploy in air-gapped environments
* Require complete control over the platform
* Operate in highly regulated industries
* Have existing on-premises infrastructure expertise
* Need custom security implementations

## Getting Started <a href="#getting-started" id="getting-started"></a>

### SaaS Deployment <a href="#saas-deployment" id="saas-deployment"></a>

1. **Sign Up**: Create your Harness account at [app.harness.io](https://app.harness.io)
2. **Create Project**: Set up your first chaos engineering project
3. **Connect Infrastructure**: Link your Kubernetes clusters or cloud accounts
4. **Run First Experiment**: Follow the [getting started guide](/resilience-testing/chaos-engineering/new-to-chaos-engineering/quickstart.md)

### Self-Managed Enterprise Edition <a href="#self-managed-enterprise-edition" id="self-managed-enterprise-edition"></a>

1. **Contact Sales**: Discuss licensing and requirements
2. **Plan Infrastructure**: Review [system requirements](/resilience-testing/chaos-engineering/use-chaos-engineering/on-premises-smp/index.md)
3. **Install Platform**: Deploy using Helm charts
4. **Configure Access**: Set up authentication and networking
5. **Connect ChaosHub**: Request Enterprise ChaosHub access if needed

## Support <a href="#support" id="support"></a>

### SaaS Support <a href="#saas-support" id="saas-support"></a>

* **Included Support**: Platform maintenance and updates
* **Community Resources**: Documentation, forums, and examples
* **Enterprise Support**: Available for enterprise customers

### Self-Managed Support <a href="#self-managed-support" id="self-managed-support"></a>

* **Installation Guidance**: Professional services available
* **Platform Updates**: Customer-managed deployment
* **Enterprise Support**: Dedicated support for enterprise customers
* **Community Resources**: Access to documentation and forums

## Next Steps <a href="#next-steps" id="next-steps"></a>

1. **Review Requirements**: Assess your security, compliance, and operational needs
2. **Compare Options**: Use the feature comparison to evaluate both deployments
3. **Plan Deployment**: Choose the option that best fits your organization
4. **Get Started**: Follow the appropriate getting started guide

For detailed information about supported platforms and limitations, see [What's Supported](/resilience-testing/chaos-engineering/new-to-chaos-engineering/whats-supported.md).