Help Center/ MaaS/ Security and Compliance/ Operation Audit and Model Monitoring
Updated on 2026-09-20 GMT+08:00
View PDF
Share

Operation Audit and Model Monitoring

Operation Audit

CTS records operations on the cloud resources in your account. You can use the logs generated by CTS to perform security analysis, trace resource changes, audit compliance, and locate faults.

After you enable CTS and configure a trace task, CTS can record management and data traces of MaaS for auditing. The event audit process is as follows.
Figure 1 CTS

For details about MaaS management and data events that can be tracked by CTS, see Table 1.

Table 1 Key MaaS operations supported by CTS

Operation

Resource Type

Trace

Creating an API key

api_key

CreateApiKey

Deleting an API key

api_key

DeleteApiKey

Updating an API key

api_key

UpdateApiKey

Signing a service statement

compliance

SignDisclaimer

Obtaining monitoring information

metrics

QueryServiceMetrics

Creating an inference service

job

CreateService

Deleting an inference service

job

DeleteService

Updating risk control settings

job

UpdateModeration

Subscribing to a model service

job

ActivateModelServices

Updating inference service configurations

job

UpdateService

Querying new-version real-time services

job

ListUserInferV2ServicesByRegions

Creating an endpoint

job

CreateCustomEndpoint

Starting or stopping an endpoint

job

ControlCustomEndpoint

Updating an endpoint

job

UpdateCustomEndpoint

Deleting an endpoint

job

DeleteCustomEndpoint
  • For details about CTS and how to enable and configure it, see CTS Overview.
  • After you enable CTS, the system records operations related to MaaS, and the console stores operation records for the past seven days by default. For details about how to view MaaS operation records, see Querying Traces in CTS.

Model Call Statistics and Monitoring

  • Model Call Statistics

    This feature collects call data of built-in services and endpoints within a specified period, including the total number of tokens called, number of input tokens, number of output tokens, and end-to-end latency. It displays data trends at a minimum granularity of minutes, helping you understand service usage and performance changes for more effective model evaluation, problem locating, fault rectification, and performance optimization. For details, go to the Call Statistics page.

  • Cloud Eye

    MaaS interconnects with Cloud Eye to monitor real-time services and their model loads, including metrics such as RPM/TPM, token usage, and model call success/failure rates. It also supports metric dashboard visualization. For details about metrics, see Metrics.

  • Simple Message Notification (SMN)

    MaaS interconnects with SMN to support alarm configuration via SMS, email, or voice for rate limiting and abnormal requests (such as error codes).

  • Run Log Recording

    The gateway data on the customer side can record logs such as service calls and token context, which can be stored in OBS as needed for process tracing.