Model-Driven Architecture 
Operations
The C3 AI Suite is a comprehensive suite of services mediated through a model-driven architecture. Platform services include data persistence, batch and stream processing, time series normalization, auto-scaling, data encryption, attribute and role-based access control, and AI/ML services.
Single sign-on
- Use any SAML v2.0 compliant identity provider
- Leverage two-factor authentication support to meet advanced authentication requirements
- Delegate user authentication and authorize third-party applications to access the C3 AI Suite using OAuth 2.0
Attribute & role-based access
- Leverage the C3 AI Suite’s comprehensive approach for managing access to data, APIs, and infrastructure resources.
- Segregate duties within the team and limit user access to resources based on job requirements.
- Define expressions to selectively provide or restrict access to application and infrastructure resources.
Auto-scaling
- Dynamically scale compute capacity up or down according to conditions you define.
- Automatically increase the number of compute instances during demand spikes to maintain performance. Decrease capacity during lulls to reduce costs.
- Maintain the cost and availability of compute resources and ensure the desired number of compute instances.
Monitoring
- Monitor all aspects of the C3 AI Suite, from the state of an analytic job to API performance.
- Visualize current and historical environment workload using an intuitive environment operations portal.
- Monitor the status and throughput of invalidation queues, the mechanism by which the C3 AI Suite handles asynchronous processing requests.
- Configure auto-scaling rules to maintain the cost and availability of compute resources.
Logging
- Track all C3 AI application access, including failed authentication attempts.
- Record all successful and unsuccessful access activities in the system and in application logs, along with identity, action, and date/time of access.
- Log every data change in the system and in application logs.
End-to-end data security
- Provide added data security using enterprise-class encryption.
- Encrypt data-at-rest using 256-bit SHA-2 encryption.
- Securely transmit data-in-motion using HTTPS (SSL/TLS) with 4096-bit RSA encryption.
Workflow
- Execute complex tasks asynchronously
- Associate workflow invocation to changes to data, user interaction, or a time-based schedule
- Monitor end-to-end execution time and status of the overall process as well as the status and execution time of each step