Installation considerations for Datalore
This page will guide through the main stages of the installation process based on your preferences and needs. The workflow below provides a choice of options. Choose the procedures that best suit you and use the links to their in-depth descriptions.
Infrastructure options
Before the installation of Datalore Enterprise itself, prepare the platform where you want to deploy it.
- On-premises
Install Datalore Enterprise on your local infrastructure.
- AWS
To deploy Datalore Enterprise on AWS, start with AWS setup. Follow the procedure here.
- Other cloud services
Below are some of the most popular cloud services you can use with the links to official docs.
Choose an installation type
The choice of the installation type depends mostly on your goals and level of knowledge and expertise.
- For testing purposes
If you are not experienced working with Kubernetes, we recommend that you go with the Docker-based installation. It's a lot quicker and simpler than the other one.
If you have experience with k8s and experience setting up variosu infrastructures, we suggest installing Datalore Enterprise on a Kubernetes cluster
- For production in use
We recommend the Kubernetes-based installation. Generally, it is required that the installation be carried out by an experienced specialist (DevOps, contract with good infrastructure-related knowledge).
Configuration procedures
Generally, the Kubernetes-based installation provides a wider range of configuration options. To configure the Datalore service after the installation, follow this procedure.
The other options are available for both installation types.
- Use JetBrains hub
You can use JetBrains Hub (further in the article referred to as Hub) for user authentication and user management in Datalore. Among other features, Hub supports most popular auth modules.
- Configure agents
Agents are virtual machines run by Datalore to perform computations. For each notebook to run, Datalore starts one more agent.
Configure agents Configure agents (Docker-based installation)
Configure agents Configure agents (Kubernetes-based installation)
- Configure plans
Configure plans to limit resources that are available for Datalore users.
- Enable gift codes and email service
With the gift codes feature enabled, you can generate or redeem gift codes that allow Datalore Enterprise users to use more computation resources or storage. The feature can be useful when a user exceeds the limits under the selected plan and needs the extension of services.
By default, emails are not enabled and email verification for users is disabled. To enable the email service, specify the following parameters for the respective installation type
- Enable user activity logging
Configure your Datalore to log the activities of your users.
(Optional) Set up external agents
With external agents, you can run Datalore computations outside your Datalore installation.
(Optional) Add Docker Swarm support
You can run your Datalore computations in a Docker Swarm, a cluster of physical or virtual machines orchestrated by the same configurations and set up to operate together.
Configure Docker Swarm agentsTroubleshooting and tips
Use the links below in case you have problems installing or configuring Datalore.
Troubleshooting
Best practices/How-to's
FAQ
VARIANT TWO
This page will guide through the main stages of the installation process based on your preferences and needs. The workflow below provides a choice of options. Choose the procedures that best suit you and use the links to their in-depth descriptions.
Installation workflow
Choose and prepare your infrastructure.
- On-premises
Install Datalore Enterprise on your local infrastructure.
- AWS
To deploy Datalore Enterprise on AWS, start with AWS setup. Follow the procedure here.
- Other cloud services
Below are some of the most popular cloud services you can use with the links to official docs.
Choose the installation type and perform the basic installation.
- For testing purposes
If you are not experienced working with Kubernetes, we recommend that you go with the Docker-based installation. It's a lot quicker and simpler than the other one.
If you have experience with k8s and experience setting up variosu infrastructures, we suggest installing Datalore Enterprise on a Kubernetes cluster
- For production in use
We recommend the Kubernetes-based installation. Generally, it is required that the installation be carried out by an experienced specialist (DevOps, contract with good infrastructure-related knowledge).
Proceed with optional configurations for the selected installation type.
- Docker-based
Integrate JetBrains Hub for user authentication and user management.
Configure agents (virtual machines run by Datalore to perform computations).
Configure plans to limit resources that are available for Datalore users.
Enable gift codes to allow users to use more computation resources and storage.
Add Docker Swarm support to run your Datalore computations in a Docker Swarm, a cluster of physical or virtual machines orchestrated by the same configurations and set up to operate together.
- Kubernetes-based
Integrate JetBrains Hub for user authentication and user management.
Configure agents Configure agents (virtual machines run by Datalore to perform computations).
Configure plans to limit resources that are available for Datalore users.
Enable gift codes to allow users to use more computation resources and storage.
Set up external agents to run Datalore computations outside your Datalore installation.
Set up external agents (Kubernetes-based installation).
Use the links below in case you have problems installing or configuring Datalore.
Troubleshooting
Best practices/How-to's
FAQ