The main Hardware Sentry dashboard provides an overview of the energy usage and carbon emissions of all data centers and server rooms. It leverages the metrics collected by Hardware Sentry OpenTelemetry Collector.
Architecture diagram: Hardware Sentry OpenTelemetry Collector runs on-prem, monitors your servers, switches and storage systems, and pushes metrics to your Datadog environment.
For each monitored host, Hardware Sentry monitors its electronic components (CPUs, memory, disks, NICs, sensors, etc.), its power consumption, and carbon emissions.
For each site (data center or server room), energy and carbon emissions are estimated for 1 day, 1 month and 1 year. Recommendation of optimal temperature is made with its potential energy savings over a year.
All hardware problems (disks, memory modules, NICs, power supplies, etc.) are monitored with specific monitors, with detailed messages.
The Hardware Sentry integration comes with a collection of recommended monitors to report hardware problems in your infrastructure.
This page is not yet available in Spanish. We are working on its translation. If you have any questions or feedback about our current translation project, feel free to reach out to us!
Overview
Hardware Sentry is an agent specialized in the monitoring of the hardware components of any server, network switch, or storage system in your data center, packaged with a collection of dashboards and monitors for Datadog.
Hardware monitoring
Hardware Sentry is a monitoring agent capable of reporting the physical health of servers, network switches, and storage systems. It collects metrics periodically to report the status of each processor, controller, disk, or power supply, the temperatures, the speed of the fans, the link status and speed of the network cards, and more.
Remote: One agent to monitor hundreds of systems, through SNMP, WBEM, WMI, SSH, IPMI, REST APIs, and more.
Multi-platform: 100+ platforms already supported with 250+ connectors (Cisco, Dell EMC, HP, Huawei, IBM, Lenovo, NetApp, Oracle, Pure, and more. For the full list of supported platforms, see the Hardware Sentry documentation.
Simple: Monitoring a system requires minimal configuration effort to specify the hostname or IP address and credentials. Hardware Sentry will automatically detect the available instrumentation and start the monitoring right away.
Normalized: All the necessary information is reported through standardized metrics in Datadog. The same hw.temperature metric, for example, is used to represent the temperature in a NetApp filer, an HP BladeSystem, a Dell PowerEdge running Windows, a Cisco UCS running Linux, or any other platform. These metrics follow OpenTelemetry’s semantic conventions.
Hardware Sentry comes with predefined monitors to detect and even predict failures in processors, memory modules, disks, network cards, controllers, power supplies, fans, temperature sensors, and more.
Energy usage and carbon footprint reports
In addition to physical health monitoring, Hardware Sentry also reports the energy usage of each monitored system. Combined with metrics representing the electricity cost and the carbon density, the provided dashboards report the electricity usage of your infrastructure in kWh and its carbon footprint in tons of CO₂.
100% Software: No smart PDUs required, even for systems that are not equipped with an internal power sensor!
The main Hardware Sentry dashboard provides an overview of the energy usage and carbon emissions of all data centers and server rooms. It leverages the metrics collected by Hardware Sentry OpenTelemetry Collector.
Architecture diagram: Hardware Sentry OpenTelemetry Collector runs on-prem, monitors your servers, switches and storage systems, and pushes metrics to your Datadog environment.
For each monitored host, Hardware Sentry monitors its electronic components (CPUs, memory, disks, NICs, sensors, etc.), its power consumption, and carbon emissions.
For each site (data center or server room), energy and carbon emissions are estimated for 1 day, 1 month and 1 year. Recommendation of optimal temperature is made with its potential energy savings over a year.
All hardware problems (disks, memory modules, NICs, power supplies, etc.) are monitored with specific monitors, with detailed messages.
The Hardware Sentry integration comes with a collection of recommended monitors to report hardware problems in your infrastructure.
