NMX Telemetry (NMX-T) Documentation

Introduction

Overview

NVIDIA NMX is a Unified Management, monitoring and analytics platform for cloud telemetry that includes several components:

  • NMX Telemetry (NMX-T): A subsystem responsible for the collection, aggregation, and transmission of telemetry data from various devices, applications, and platforms within a data center.

  • NMX Manager (NMX-M): An event-driven microservice architecture system that collects and processes data center telemetry, providing monitoring, insights, and predictions on the operability and health of systems. It can aggregate, analyze, run machine learning models for inference and pattern detection, and control the behavior of the HCP by changing the configuration of network or compute entities using the NMX Controller.

  • NMX Controller (NMX-C): A control plane entity responsible for the configuration, monitoring, and control of various systems, mainly network devices, within the data center.

  • NMX Oasis: A data lake solution that resides in single or multiple clouds, consisting of API gateways, ETL processes, compute clusters, analysis models, and informative dashboards. The suite aims to provide a comprehensive solution for managing and analyzing telemetry data to optimize the performance and health of data center operations.

NMX Solution topology

NMX Solution Topology.png

Application Layout

NMX-T Application Layout

NMX-T layout-NMX-T v0.8 layout.png

Interfaces

List of the application interfaces, including the internal ones. Integration parameters are managed as described Installation Configuration

Port

Description

integration.json parameter

Scope

Further information

9350

Control interface

control_port

Host-internal

Interface Configuration

9351

gRPC-based Connection API

grpc_port

Public

gRPC Interface

9352

Prometheus metrics endpoint

prometheus_exporter_port

Public

Prometheus Metrics Endpoint

9353

syslog aggregation service 

syslog_receiver_port

Public

Collected Data | id (1.2.3 GB200 1.3)CollectedData .Collecteddatav1.0.0GA Syslogmessages

9360

IB Telemetry metrics interface

n/a

App internal

Internal interface, to be hidden away in upcoming releases.

9161

IB Telemetry Sink interface 

n/a

App internal

Internal interface, to be hidden away in upcoming releases.


Last updated: