1.9 KiB
Installation Overview
The Stoplight Enterprise platform provides a fully-functional on-premise API design, test, and documentation toolkit, taking the hassle out of your API strategy.
Deployment Options
Before beginning with the installation, be sure to prepare all of the necessary systems that will be running Stoplight components. For more information on sizing, disk, and network requirements, please see the Technical Requirements section.
Single-server vs. Multi-server Deployments
Stoplight can be deployed on one or many Linux servers (dedicated or virtualized). Single-server deployments run all of the necessary Stoplight components on a single Linux instance. This greatly simplifies the deployment process, as all components do not have to reach over the network to talk to one another. Despite ease of installation, there are some notable shortcomings to this option:
-
If the system is taken down for any reason, all components will be unavailable.
-
Any single component can affect the performance of the entire Stoplight platform, leading to service degradation across all components.
Due to the shortcomings listed above, single-server deployments are typically recommended for POC/trial environments or for smaller organizations that do not wish to allocate multiple servers for the Stoplight platform.
Multi-server deployments run different Stoplight Enterprise components on separate Linux instances. This deployment option is much more resilient to system-level issues, though it does require more network configuration.
Stoplight recommends multi-server deployments for all production environments.
Native vs. Container-based Deployments
The Stoplight platform can be run either with a container solution (Docker) or natively on the Linux system. Both options are supported, however Stoplight recommends leveraging containers where possible for ease-of-use and improved security/sandboxing.