Runtime Metrics
Runtime metrics are available for runtime implementations that expose metric reporting.
The following tags are present on all runtime metrics:
The following tags are present on all tokio runtime metrics:
runtime_id
The runtime ID or label.
There may be multiple instances of the same runtime type. This tag distinguishes between them.
Jemalloc Runtime Metrics
There metrics come from the jemalloc memory allocator.
runtime.jemalloc.approximate_active
type: gauge
The total number of bytes in active pages collected in an unsynchronized manner, without requiring an epoch update.
Added in version 0.4.0.
Mimalloc Runtime Metrics
There metrics come from the mimalloc memory allocator.
runtime.mimalloc.current_pages
type: gauge
Current count of mimalloc pages.
runtime.mimalloc.peak_pages
type: gauge
Peak count of mimalloc pages.
runtime.mimalloc.current_commit
type: gauge
Current committed memory (backed by the page file).
runtime.mimalloc.peak_commit
type: gauge
Peak committed memory (backed by the page file).
Added in version 0.4.0.
Tokio Runtime Metrics
These metrics come from the Tokio runtime.
The supported metrics are:
runtime.tokio.alive_tasks
type: gauge
Current number of live tasks in the runtime.
runtime.tokio.global_queue_depth
type: gauge
Number of tasks currently scheduled in the runtime’s global queue.