10 General Considerations for Container Configuration

When you configure the Private AI Services Container, there are a number of options and information to consider that can help you get the most out of this service.

• Rate Limiting
  The Private AI Services Container provides a configurable method to control the number of requests per minute that the container can handle for different categories of endpoints. This improves scalability and helps prevent abuse, including denial-of-service (DOS) attacks.
• HTTP Request Input Validation
  The REST API accepts the media type application/json for all requests that have a body (payload), and returns application/json for all responses.
• Use API Keys in Clients
  Clients to the Private AI Services Container must send a valid API key if the service has PRIVATE_AI_AUTHENTICATION_ENABLED set to true.
• Support for Multiple API Keys
  Multiple API keys can be associated with the container at startup as a means of authentication. The use of multiple API keys enables finer-grained access management and improved auditing and attribution of requests
• Transport Layer Security
  The container admin can manage transport layer security using environment variables and a keystore file, along with a password file to access the keystore.
• Container Log Files
  Diagnostic information about the Private AI Services Container is available and stored in log messages.