MetricsHub® Enterprise: Insights from the Team Behind the Innovation

After months of intense development, our team is proud to unveil MetricsHub® Enterprise. This major milestone in the evolution of our platform gives us a perfect opportunity to reflect on the journey so far, with insights from Nassim, our Software Architect.

Hardware monitoring as the foundation

MetricsHub® might seem new, but it builds on the robustness of a hardware monitoring solution with over 20 years of proven success: Hardware Sentry. Originally developed for BMC Software’s users, the solution was completely rewritten in 2022 to embrace OpenTelemetry, the new standard for observability. After years of serving BMC customers, the solution gained broader industry recognition, integrating into the Datadog Marketplace and being featured at GrafanaCon 2023, where our CEO, Bertrand Martin, showcased how Sentry Software reduced energy consumption with Hardware Sentry and Grafana.

Going beyond hardware monitoring…

As we gained expertise in OpenTelemetry, we recognized that the core engine of Hardware Sentry held untapped potential, capable of doing much more than just hardware monitoring. We envisioned a flexible, powerful solution capable of monitoring any resource, whether within private enterprise networks or across both public and hybrid cloud environments. This is how the project MetricsHub® started.

The strength of Hardware Sentry lies in its extensive library of connectors, which were designed to collect hardware information from servers, storage systems, and network switches.

Laying the foundation for great product development

Value #1: Learning from experts

Claiming to know everything would be presumptuous. Acknowledging that no one can master everything, we actively collaborate with industry experts and academic institutions. We engage with developer communities and have established technical partnerships with leading observability platform vendors such as Datadog, Splunk, New Relic, Grafana, and Prometheus.

Value #2: Developing with the customer in mind

Customer feedback has played a significant role in shaping MetricsHub®. By collaborating closely with our customers, we gained invaluable insights into their needs and crafted a solution tailored to their requirements. For instance, a major telecom company required several specific customizations: the ability to add new metrics, to easily adjust collection times, and to scale the solution to monitor numerous resources without disrupting data collection. Their feedback encouraged us to go open source to increase transparency and build trust in the platform through contributions.

The MetricsHub project team works around a table to design the software solution's functionalities

Staying updated on industry trends, emerging technologies, and evolving customer needs over secondary features is vital for integrating the latest innovations into Metricshub® and ensuring it evolves alongside market demands. Our technical research includes analyzing market reports, attending industry conferences and engaging with thought leaders. AI, machine learning, and observability standards rank among the key areas that drive our development efforts. This proactive approach keeps MetricsHub® at the forefront of innovation, addressing the complex needs of modern IT environments.

Overcoming significant development challenges

Challenge #1: Converting proprietary connectors to open formats

The strength of Hardware Sentry lies in its extensive library of connectors, which were designed to collect hardware information from servers, storage systems, and network switches. However, these connectors were originally written in a proprietary language and stored in Hardware Definition Files (HDF). To ensure compatibility with MetricsHub®, we converted these connectors into YAML format. We adopted a Test-Driven Development (TDD) approach to ensure safety, reliability and backward compatibility, which enabled us to run the Hardware Sentry ancestor solution with YAML connectors. This success laid a solid foundation for MetricsHub®.

Challenge #2: Finding the most effective configuration method

To offer a straightforward yet powerful configuration solution, we designed a file that sets up the core features of MetricsHub® while allowing advanced customizations (metricshub.yaml). Users can create custom connector snippets to enhance the data collection process and enrich the collected metrics.

A young IT technician works on her laptop in front of a data center

Challenge #3: Adapting the core engine

The core engine, originally designed for hardware monitoring, needed to be adapted to a more generic model. Unlike Hardware Sentry, which was highly specialized, MetricsHub® needed to be broadly applicable to monitor physical, logical, or virtual resources alike. This required a redesign of our internal model.

Challenge #4: Breaking Down Features: Community vs. Enterprise Editions

Differentiating the features between the Community and Enterprise Editions was a challenging task. We aimed to offer meaningful functionality to the Community version while reserving advanced features for Enterprise users. This presented challenges not only from a functional perspective but also from a technical one. To achieve this balance, we conducted several internal working sessions and consulted experts, including Gartner®. We believe we have successfully struck a balance that will meet the needs and expectations of both our Community and Enterprise customers.

Looking to the future…

We have exciting plans to make future versions of MetricsHub® even better! Some of the features on our roadmap include a user interface, expanded support for additional resources like storage systems, databases, and applications, and integrations with more observability platforms. Want to stay updated with our latest developments? Join our Slack channel and be the first to know!

Join our Slack channel Talk to an expert

Share this post