Enterprise evaluation FAQ

• Architecture

  • Data flows
  • Deployment

• Security

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

• Enterprise Readiness

  • Logging and monitoring (using standard AWS tools)
  • 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 and integrate that data with Datagrok API
  • Interacting with Datagrok
    • server API
    • client API
  • Connecting to common data sources
    • relational databases
    • local files
    • datastore files
  • 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
  • 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

  • Capacity of holding data in client with differently sized data sets related to project scenarios, sizes detailed in data sources section below
    • High-throughput screening
    • Virtual screening (minimal requirement)
    • Large-chemical spaces
  • Ability to cross-connect multiple large data tables
  • Visualization of data sets in different plots
    • High-throughput screening
    • Virtual screening
  • 2D | 3D structure rendering and interaction:
    • Ability to customize structure rendering, and interact with it
    • Atom-based annotations
    • 2D-3D connectivity
  • Interactivity
    • Live data masking
    • Filter by selection
    • Different input methods (2D drawing etc.)
    • API calls
  • Ability to plugin non-native Datagrok pieces (e.g. react containers) and interact with Datagrok frontend (eg react containers)
  • Scalability of frontend scripting functionality

Encryption at rest

For AWS deployment, we rely on Amazon's built-in encryption for RDS and S3 buckets.

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

For AWS deployment, we rely on Amazon CloudWatch service which provides a convenient way to collect and observe metrics and logs.

Backup and restore

Amazon has scheduled backups for RDS and S3, but you can back up and restore an RDS database as a standard Postgres database.

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.

Docker-compose manifest is used to describe and deploy Datagrok applications.

Also, there are multiple advanced options to deploy application:

• CloudFormation template to deploy to AWS ECS
• Terraform scripts to deploy to AWS ECS