Vector Database Metrics
Vector database metrics in this repo are for the Milvus metrics path. The ready-to-use exporter maps Milvus core metrics, proxy operation metrics, cache-hit metrics, and queue-latency metrics into AppDynamics custom metrics for the vector database dashboard.
Prerequisites
- Milvus is deployed and exposing Prometheus metrics
- service path is reachable from the AppDynamics scrape path
- Prometheus-compatible metrics are available at
/metrics
Enable Prometheus Scraping for Vector Database
- service:
milvus-deployment - namespace:
milvus - port:
9091 - path:
/metrics
Replace these values with the vector database metrics service name and namespace used in the target environment.
Configure Machine Agent Ingestion
Infrastructure Visibility Prometheus monitoring loads the vector database exporter definition through prometheus-config-template.yamlNo Content found for /db/organizations/splunk/repositories/appdsaas-26_4_0/content/documents/AppDynamics/attachments/custom_dashboards_ai_pods/reference_configs_ai_pods.zip.
To activate this path:
- include the vector database exporter YAML in the
prometheus-exportersConfigMap - ensure that the main include list references that exporter file
- apply the ConfigMap update and re-run the Infrastructure Visibility Helm upgrade
Before enabling the scrape, update the exporter YAML service discovery fields to the service name and namespace used by your vector database deployment.
Exporter YAML Contract
- exporter-yamls/vector-database-core-metrics.yamlNo Content found for /db/organizations/splunk/repositories/appdsaas-26_4_0/content/documents/AppDynamics/attachments/custom_dashboards_ai_pods/exporter_yamls_ai_pods.zip
- key source metrics:
milvus_querycoord_collection_nummilvus_rootcoord_dml_channel_nummilvus_rootcoord_id_alloc_countmilvus_proxy_req_countmilvus_proxy_cache_hit_countmilvus_rootcoord_ddl_req_countmilvus_proxy_req_in_queue_latency_summilvus_proxy_req_in_queue_latency_countmilvus_rootcoord_ddl_req_latency_in_queue_summilvus_rootcoord_ddl_req_latency_in_queue_count
- custom metric family:
Custom Metrics|Vector DB|...
Expected AppDynamics Custom Metric Paths
Custom Metrics|Vector DB|QueryCoord Collection CountCustom Metrics|Vector DB|RootCoord DML Channel CountCustom Metrics|Vector DB|RootCoord ID Alloc CountCustom Metrics|Vector DB|Proxy|{status}|Request CountCustom Metrics|Vector DB|Proxy Request CountCustom Metrics|Vector DB|Proxy|Cache|{cache_state}|Hit CountCustom Metrics|Vector DB|Proxy|Cache|{cache_name}|Hit Count (agg)Custom Metrics|Vector DB|Proxy|{function_name}|Request Count (agg)Custom Metrics|Vector DB|RootCoord|{function_name}|DDL Request Count (agg)Custom Metrics|Vector DB|Proxy|{function_name}|Avg Latency in Queue (ms) (agg)Custom Metrics|Vector DB|RootCoord|{function_name}|Avg DDL Latency in Queue (ms) (agg)
Create Custom Dashboard
The custom dashboard script generates ready-to-import AppDynamics dashboard JSON files from a set of templates. You supply your environment's node names and, optionally, the custom metric path prefixes. The script substitutes them into the templates and writes the JSON files. See Create Custom Dashboards for AI Pods.
Troubleshooting
- if queue-latency widgets are empty, verify that both
_sumand_counthistogram series are present upstream. - distinguish the zero values from missing paths before changing the exporter YAML.