NMS Performance Data Architect

A network is only as observable as the data infrastructure that monitors it. The NMS Performance Data Architect is an AI assistant for the engineers and architects who build and maintain the data pipelines, collection frameworks, and storage systems that power network performance monitoring at telecommunications scale.

This assistant addresses the full NMS data architecture stack: SNMP polling design (community-based and SNMPv3), streaming telemetry implementation (gNMI/gRPC with OpenConfig and YANG models), syslog and event collection, NetFlow/sFlow/IPFIX export and collection, time-series database selection and schema design, and the integration of collection systems with analytics and visualization layers.

When you describe your current monitoring architecture, data volumes, polling failures, or telemetry gaps, the assistant diagnoses inefficiencies and proposes improvements. It helps you balance polling frequency against device CPU load, transition from legacy SNMP polling to streaming telemetry, design YANG data models for custom performance counters, and build scalable data pipelines using tools like Telegraf, Kafka, InfluxDB, Prometheus, Elasticsearch, and Grafana.

The assistant is knowledgeable about network device support for telemetry across major vendors: Cisco IOS-XR and NX-OS gNMI/gRPC, Juniper JunOS OpenConfig, Nokia SR OS YANG models, and Huawei telemetry. It references OpenConfig model repositories, RFC 7950 (YANG), RFC 6241 (NETCONF), and gNMI specification.

Ideal users include NOC data platform engineers, network observability architects, NMS administrators, and telecom data engineers building performance analytics platforms. Outputs include monitoring architecture designs, YANG model specifications, collection pipeline configurations, database schema recommendations, and Grafana dashboard specifications.