vllm.config.observability ¶
ObservabilityConfig ¶
Configuration for observability - metrics and tracing.
Source code in vllm/config/observability.py
collect_detailed_traces class-attribute
instance-attribute
¶
collect_detailed_traces: Optional[
list[DetailedTraceModules]
] = None
It makes sense to set this only if --otlp-traces-endpoint
is set. If set, it will collect detailed traces for the specified modules. This involves use of possibly costly and or blocking operations and hence might have a performance impact.
Note that collecting detailed timing information for each request can be expensive.
collect_model_execute_time cached
property
¶
collect_model_execute_time: bool
Whether to collect model execute time for the request.
collect_model_forward_time cached
property
¶
collect_model_forward_time: bool
Whether to collect model forward time for the request.
otlp_traces_endpoint class-attribute
instance-attribute
¶
Target URL to which OpenTelemetry traces will be sent.
show_hidden_metrics cached
property
¶
show_hidden_metrics: bool
Check if the hidden metrics should be shown.
show_hidden_metrics_for_version class-attribute
instance-attribute
¶
Enable deprecated Prometheus metrics that have been hidden since the specified version. For example, if a previously deprecated metric has been hidden since the v0.7.0 release, you use --show-hidden-metrics-for-version=0.7
as a temporary escape hatch while you migrate to new metrics. The metric is likely to be removed completely in an upcoming release.
__post_init__ ¶
Source code in vllm/config/observability.py
_parse_collect_detailed_traces ¶
compute_hash ¶
compute_hash() -> str
WARNING: Whenever a new field is added to this config, ensure that it is included in the factors list if it affects the computation graph.
Provide a hash that uniquely identifies all the configs that affect the structure of the computation graph from input ids/embeddings to the final hidden states, excluding anything before input ids/embeddings and after the final hidden states.