Skip to content

Architecture

The image below outlines the architecture and approach used by Rafay to centrally aggregate GPU and VM telemetry from end user virtual machines (VMs) hosted in the cloud provider's datacenter. The metrics are aggregated using OpenTelemetry (OTel) and synchronized to a centralized Time Series Database at the Rafay Controller for end-user visualization and analytics.

Metrics Flow Architecture

Info

Only Ubuntu 22.04 and 24.04 OS based VMs are currently supported for integrated metrics.


Key Capabilities

Open Standards

Uses OpenTelemetry for portability and extensibility

Multi-VM Scaling

Architecture supports deployment across many VMs

Tenant Isolation

Metrics collected per-host, enabling multi-tenant observability

No Kernel Modules

All exporters and collectors run in user space


Central Time Series Database

Metrics data from all the VMs under management is aggregated in a time series database co-located on the Rafay Controller. This centralized telemetry backend:

  1. Stores time-series data from all VMs
  2. Supports querying and dashboard rendering
  3. Implements retention and downsampling policies

Components

The following modules are used for metrics aggregation at the Virtual Machine.

NVIDIA DCGM Exporter

This component collects GPU-related metrics such as:

  • Memory utilization
  • Core utilization
  • Temperature
  • Health status

Host Metrics Exporter

This component collects VM/system-level metrics including:

  • CPU usage
  • Memory usage
  • Disk I/O
  • Network statistics

OpenTelemetry (OTel) Collector

This is operated as a local service on the VM.

  • It scrapes data from the DCGM and Host Metrics exporters
  • Normalizes and prepares metrics
  • Forwards metrics to the central TSDB at the Rafay Controller

Important

By default, metrics data is aggregated from the VM to the centralized TSDB every 60 seconds.