Lambda Functions - Amazon CloudWatch

Lambda Functions

The following metrics are available as time series aggregated data in CloudWatch Metrics in the LambdaInsights namespace for standard Lambda functions.

Metric name Dimensions Description

cpu_total_time

function_name

function_name, version

Sum of cpu_system_time and cpu_user_time.

Unit: Milliseconds

init_duration

function_name

function_name, version

The amount of time spent in the init phase of the Lambda execution environment lifecycle.

Unit: Milliseconds

memory_utilization

function_name

function_name, version

The maximum memory measured as a percentage of the memory allocated to the function.

Unit: Percent

used_memory_max

function_name

function_name, version

The measured memory of the function's execution environment.

Unit: Megabytes

rx_bytes

function_name

function_name, version

The number of bytes received by the function.

Unit: Bytes

tmp_free

function_name

function_name, version

The amount of space available in the /tmp directory.

Unit: Bytes

tmp_used

function_name

function_name, version

The amount of space used in the /tmp directory.

Unit: Bytes

tx_bytes

function_name

function_name, version

The number of bytes sent by the function.

Unit: Bytes

total_memory

function_name

function_name, version

The amount of memory allocated to your Lambda function. This is the same as your function’s memory size.

Unit: Megabytes

total_network

function_name

function_name, version

Sum of rx_bytes and tx_bytes. Even for functions that don't perform I/O tasks, this value is usually greater than zero because of network calls made by the Lambda runtime.

Unit: Bytes

used_memory_max

function_name

function_name, version

The measured memory of the function sandbox.

Unit: Megabytes

The following metrics can be found in the embedded metric format log entries by using CloudWatch Logs Insights. For more information about CloudWatch Logs Insights, see Analyzing Log Data with CloudWatch Logs Insights.

For more information about embedded metric format, see Embedding metrics within logs.

Metric name Description

agent_version

The current version of the Lambda Insights extension agent running on the Lambda function.

Unit: String

cpu_user_time

The amount of time the CPU spent executing user code.

Unit: Milliseconds

cpu_system_time

The amount of time the CPU spent executing kernel code.

Unit: Milliseconds

cpu_total_time

Sum of cpu_system_time and cpu_user_time.

Unit: Milliseconds

fd_use

File descriptors currently in use.

Unit: Count

fd_max

Maximum file descriptors available for use.

Unit: Count

version

The Lambda function's version for which the other metrics were collected.

Unit: Count

agent_memory_max

Maximum memory consumed by the Lambda Insights extension agent.

Unit: Bytes

agent_memory_avg

Average memory consumed by the Lambda Insights extension agent.

Unit: Bytes

memory_utilization

The average memory measured as a percentage of the memory allocated to the execution environment.

Unit: Percent

used_memory_max

The measured memory of the function's execution environment.

Unit: Megabytes

rx_bytes

The number of bytes received by the function.

Unit: Bytes

tx_bytes

The number of bytes sent by the function.

Unit: Bytes

threads_max

The number of threads in use by the function process. As a function author, you don't control the initial number of threads created by the runtime.

Unit: Count

tmp_used

The amount of space used in the /tmp directory.

Unit: Bytes

tmp_max

The amount of space available in the /tmp directory.

Unit: Bytes

total_memory

The amount of memory allocated to your Lambda function. This is the same as your function’s memory size.

Unit: Megabytes

total_network

Sum of rx_bytes and tx_bytes. Even for functions that don't perform I/O tasks, this value is usually greater than zero because of network calls made by the Lambda runtime.

Unit: Bytes