Boomi Agent Garden provider: Monitoring agents
The metrics for the Boomi Garden agents are available at the first sync after provisioning your account. These metrics are updated at every session that an agent is invoked. The following API schema corresponds to the metrics being pulled:
| Metric name | Description |
|---|---|
| ModelInvocationCount | Counts the number of times the model has been invoked, providing insight into runtime activity. |
| ModelLatency | Measures the average time taken for a process or operation, indicating overall performance. |
| InputTokenCount | Measures the number of input tokens processed, indicating how much data is being used. |
| OutputTokenCount | Measures the number of output tokens generated, reflecting the volume of data produced by the model. |
| InvocationClientError | All the client errors during invocation. |
| InvocationServerErrors | Counts server-side errors during invocation, indicating Runtime issues. |
| ModelInvocationServerErrors | Similar to InvocationServerErrors but specific to model invocations. |
| ModelInvocationThrottles | Throttling responses received specific to the model invocations. |
| ModelInvocationClientErrors | Similar to InvocationClientErrors but specific to model invocations. |
| ModelInvocationUnknowsError | Errors which are not classified specific to the model invocations. |
| TTFT | Time to first token while invocation. |
| TotalTime | Total time taken to complete the single invocation response. |
| Time | Time stamp of the event. |
| ExtAccountId | Boomi account is specific to Boomi platform. |
| ProviderType | Here provider type will always be Boomi. |
| Operation | Operation happening like invoking the agent. |
| ExtAgentId | ID of the agent that is being invoked. |
| SchemaVersion | Version of metric schema being sent to control tower. |
To know about filtering through the observed metrics, refer to Monitoring and Compliance.
