Based on the resource structure and scenarios, Function Compute divides monitoring metrics by region, service, and function. These monitoring metrics correspond to the input parameter Metric in requests for calling the Function Compute API. This topic describes the metrics for each dimension.

Region-specific metrics

Region-specific metrics are used to monitor the usage of Function Compute resources in a region. The following table lists the region-specific metrics.

Metric alias Unit Description
TotalInvocations Count The total number of requests to invoke functions in the region, including:
  • All requests that call the InvokeFunction operation, except those for which the HTTP 202 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.
    Note A request for an asynchronous invocation is not counted in this metric when the HTTP 202 status code is returned, but is counted only when the system attempts to execute the invoked function.
BillableInvocations Count The number of requests for successful invocations in the region, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 200 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.

Same as the requests counted in the billing parameter Invocations, the invocation requests recorded in this metric are billable.

Note The preceding invocation requests are counted in this metric even if function execution fails due to syntax or execution errors.
Throttles Count The number of requests for invocations in the region followed by function execution failures caused by throttling, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by throttling.
ClientErrors Count The number of requests for invocations in the region followed by function execution failures caused by client errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 4xx status code except the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by client errors.
ServerErrors Count The number of requests for invocations in the region followed by function execution failures caused by server errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 5xx status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by server errors.
BillableInvocations percentage % The percentage of requests for billable invocations in the region.
Throttles percentage % The percentage of requests for invocations in the region followed by function execution failures caused by throttling.
ClientErrors percentage % The percentage of requests for invocations in the region followed by function execution failures caused by client errors.
ServerErrors percentage % The percentage of requests for invocations in the region followed by function execution failures caused by server errors.
RegionOnDemandUsage GB*s Pay-as-you-go resource usage of each account.
RegionInvokeInternetOut Bit Internet traffic generated to respond to invocations in each account.
RegionDataTransferInternetOut Bit Internet traffic generated for Internet access by functions in each account.

Service-specific metrics

Service-specific metrics are used to monitor the resource usage of a service. The following table lists the service-specific metrics.

Metric alias Unit Description
TotalInvocations Count The total number of requests to invoke functions in the service, including:
  • All requests that call the InvokeFunction operation, except those for which the HTTP 202 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.
    Note A request for an asynchronous invocation is not counted in this metric when the HTTP 202 status code is returned, but is counted only when the system attempts to execute the invoked function.
BillableInvocations Count The number of requests for successful invocations in the service, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 200 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.

Same as the requests counted in the billing parameter Invocations, the invocation requests recorded in this metric are billable.

Note The preceding invocation requests are counted in this metric even if function execution fails due to syntax or execution errors.
Throttles Count The number of requests for invocations in the service followed by function execution failures caused by throttling, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by throttling.
ClientErrors Count The number of requests for invocations in the service followed by function execution failures caused by client errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 4xx status code except the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by client errors.
ServerErrors Count The number of requests for invocations in the service followed by function execution failures caused by server errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 5xx status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by server errors.
BillableInvocations percentage % The percentage of requests for billable invocations in the service.
Throttles percentage % The percentage of requests for invocations in the service followed by function execution failures caused by throttling.
ClientErrors percentage % The percentage of requests for invocations in the service followed by function execution failures caused by client errors.
ServerErrors percentage % The percentage of requests for invocations in the service followed by function execution failures caused by server errors.
ServiceOnDemandUsage GB*s Pay-as-you-go resource usage of the service.
ServiceInvokeInternetOut Bit Internet traffic generated to respond to invocations in the service.

Function-specific metrics

Function-specific metrics are used to monitor the resource usage of a function. The following table lists the function-specific metrics.

Metric alias Unit Description
Average duration ms Request duration measures the length of time that elapsed from the start to the end of function execution after the function is invoked. Average duration averages the duration of all requests for the function, in minutes.
Note The billing system rounds the request duration up to the nearest 100 ms. Here, Function Compute calculates the actual execution time.
Highest memory usage MB The largest amount of memory that is used when the function is invoked during the aggregation period.
TotalInvocations Count The total number of requests to invoke the function, including:
  • All requests that call the InvokeFunction operation, except those for which the HTTP 202 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.
    Note A request for an asynchronous invocation is not counted in this metric when the HTTP 202 status code is returned, but is counted only when the system attempts to execute the invoked function.
BillableInvocations Count The number of requests for successful invocations of the function, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 200 status code is returned.
  • Requests for asynchronous invocations followed by function execution attempts.

Same as the requests counted in the billing parameter Invocations, the invocation requests recorded in this metric are billable.

Note The preceding invocation requests are counted in this metric even if function execution fails due to syntax or execution errors.
Throttles Count The number of requests for invocations of the function followed by function execution failures caused by throttling, including:
  • Requests that call the InvokeFunction operation and for which the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by throttling.
FunctionErrors Count The number of requests for invocations of the function followed by function execution failures caused by function execution errors, including:
  • HandledInvocationError: errors returned when a callback function is executed in Node.js. For more information, see Handle errors.
  • UnhandledInvocationError: all errors except those recorded in HandledInvocationError. For more information, see Handle errors.
  • OOM: errors returned because of out-of-memory.
  • Timeout: errors returned because of function execution timeout.
ClientErrors Count The number of requests for invocations of the function followed by function execution failures caused by client errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 4xx status code except the HTTP 429 status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by client errors.
ServerErrors Count The number of requests for invocations of the function followed by function execution failures caused by server errors, including:
  • Requests that call the InvokeFunction operation and for which an HTTP 5xx status code is returned.
  • Requests for asynchronous invocations followed by function execution failures caused by server errors.
BillableInvocations percentage % The percentage of requests for billable invocations of the function.
Throttles percentage % The percentage of requests for invocations of the function followed by function execution failures caused by throttling.
FunctionErrors percentage % The percentage of requests for invocations of the function followed by function execution failures caused by function errors.
ClientErrors percentage % The percentage of requests for invocations of the function followed by function execution failures caused by client errors.
ServerErrors percentage % The percentage of requests for invocations of the function followed by function execution failures caused by server errors.
FunctionOnDemandUsage GB*s Pay-as-you-go resource usage of the function.
FunctionInvokeInternetOut Bit Internet traffic generated to respond to invocations of the function.
Note The average duration metric measures the average of the recorded values, in minutes. All other metrics measure the sum of the recorded values, in minutes.