Configure AI Components to Collect Metrics

This page describes how to configure AppDynamics to collect data from the following AI components:

Cisco Unified Computing System Server Metrics

Cisco AI PODs are integrated with Cisco Unified Computing System (UCS) servers. You can monitor the performance of Cisco UCS servers with Cisco Intersight, an IT operation and server management platform. See Cisco Intersight Metrics.

Cisco Nexus Metrics

Cisco AI PODs are integrated with Cisco Nexus data center switches. You can collect Cisco Nexus metrics using the Cisco OS receiver, which exports metrics from Cisco devices through an SSH connection. See Cisco Nexus Metrics.

NVIDIA GPU Metrics

You can monitor the performance of NVIDIA GPUs by configuring your Kubernetes cluster to send NVIDIA GPU metrics to Splunk AppDynamics. This solution uses the Prometheus receiver to collect metrics from the NVIDIA DCGM Exporter, which can be installed independently or as part of the NVIDIA GPU Operator. See NVIDIA GPU Metrics.

NVIDIA NIM Metrics

You can monitor the performance of NVIDIA NIMs by configuring your Kubernetes cluster to send NVIDIA NIM metrics to Splunk AppDynamics. See NVIDIA NIM Metrics.

Pure Storage Metrics

Cisco AI PODs are designed to integrate with high-performance storage solutions. You can monitor the performance of Cisco AI PODs that use Pure Storage with Portworx. Portworx is a Kubernetes data services platform that provides persistent storage, data protection, and disaster recovery for containerized applications. See Pure Storage Metrics.