Enterprise evaluation FAQ

• Architecture

  • Infrastructure
  • Deployment

• Security

  • Security, authentication, and authorization
  • Encryption at rest and encryption in transit
  • Vulnerability remediation

• Enterprise readiness

  • Managing an enterprise instance
  • Logging and monitoring
  • Backup and restore
  • Disaster recovery (HA/DR)
  • Infrastructure as code (ability to deploy using standard DevOps tools)

• Interoperability

  • Calling web services from client and server with proper auth: OpenAPI
  • Interacting with Datagrok
    • server API
    • client API
  • Connecting to common data sources
    • relational databases
    • local files
    • file shares and cloud storage
  • Embedding a custom visualization into Datagrok

• Developer experience

  • Debug environment when developing customizations
  • Devops process including deployment of packages, their dependencies and versioning
  • Concurrent work by team of developers

• Scalability and performance

  • Maximum dataset sizes and in-memory performance
  • Stability under concurrent user load
  • Scaling and stability under load

• Extensibility

  • Creating custom visualizations
  • Creating custom server-side components
  • Creating custom scripts and utilizing them in other components
  • Ability to reskin Datagrok to appear as a fit-for-purpose web application
  • Ability to build custom application including data entry, workflow, data model, state management, persistence, etc

• Frontend

  • Holding and exploring large datasets in the browser
  • Visualizing datasets with interactive viewers
  • 2D and 3D structure rendering, sketching, and search
  • 3D biostructures
  • Interactivity
    • Live data masking
    • Filter by selection
  • Developing custom viewers, including non-native components such as React containers

Encryption at rest

For AWS deployment, we rely on Amazon's built-in encryption for RDS and S3 buckets. Credentials for data connections are additionally encrypted by Datagrok with server keys.

Encryption in transit

All client-server communications use the HTTPS protocol, which means it is secure and encrypted.

Server API

The Datagrok client uses an HTTP REST API to interact with the server. You must pass an authentication token to access all features. Proof of concept video

Logging and monitoring

Datagrok works with the monitoring tools you already use, in any cloud or on-premises:

• Logs and alerts. Log sync pushes logs, audit events, alerts, and a heartbeat to Amazon CloudWatch, Google Cloud Logging, or any OpenTelemetry (OTLP) collector. See Export logs.
• Health checks. The /api/admin/health endpoint reports the status of every service and needs no sign-in, so load balancers and uptime monitors can probe it.
• Usage and performance. Usage Analysis shows user activity, errors, and server and database metrics.

For details, see Monitor the platform.

Backup and restore

Back up these together so that you can restore them to the same point in time:

• The Postgres database that holds metadata, audit logs, and encrypted credentials. You can back it up and restore it as a standard Postgres database, for example with scheduled RDS snapshots on AWS.
• The file storage (S3, Google Cloud Storage, or a local volume). On AWS, see S3 backup.
• Server key material, if you keep keys in an external backend.

Disaster recovery

Datagrok supports Docker installation, Amazon cluster will immediately restart failed instance

Infrastructure as a Code

Datagrok Docker containers are built using Jenkins. All software is upgraded and patched on every build.

You can deploy Datagrok with standard DevOps tools:

• Docker Compose
• Helm chart for Kubernetes
• CloudFormation template for Amazon EKS
• Terraform for AWS
• Terraform for Google Cloud (GKE)