Monitoring IT Operations in the Datacenter

The question for those in data center operations management is not which tool but rather what combination of tools will provide the right metrics at the right time.  Quest solutions are part of the larger IT ecosystem and thus our products tend to integrate well with other complimentary and even competitive solutions.  Foglight is one such product.  It is often times made the record of truth for enterprise performance monitoring while feeding or pulling from other tools to provide the needed intelligence toward preventing system failures.  With Foglight administrators can manage by exception thus culling the alarm noise down to the essential need to know metrics.  This simply means knowing what is wrong versus a constant stream of events driving alerts which in turn demand action.  Foglight’s centralized dashboards provide a single view of system and application alerts from all levels of the IT architecture. 

 Any enterprise monitoring team managing complex systems will need to understand three core data points about the target platforms:

  • The applications architecture: Whether single compiled binary, JVM/Java, .Net/JIT, client/server, n-tier etc. Here Foglight allows visibility into all tiers of an applications structure while showing dependencies in a topology view
  • The workloads configuration: Physical OS, virtual hypervisor, private/public cloud, container (Docker) etc. Here Foglight’s extensible model provides options to directly monitor workloads regardless of the deployed configuration
  • The consumption baseline: Foglight uses IntelliProfile to evaluate collected data against the baseline, and compares incoming data for those metrics that have IntelliProfile threshold levels configured. Metric threshold states reflect the degree of deviation from the baseline, and can indicate potential performance bottlenecks. If there are any rule conditions that evaluate threshold states for such metrics, Foglight can generate alarms when a metric enters a certain threshold state. These automatically computed baselines for all monitored workloads provide a consistent window into performance metrics the administrator decides to correlate.  This makes apple to apple comparisons readily accessible to all system stakeholders.

These three cornerstones provide the context from which efficient operational intelligence can be produced and used proactively.  The alternative is alert storms, metric bombardment and a general atmosphere of firefighting within the data center.  Context also reduces the long running saga of finger pointing across areas of responsibility within operations.  This allows you to maximize team collaboration based on a common platform and single version of the truth.

About the Author
Chris Roberts
Tinker, Maker, Coder, Builder, Creator and fan of all things technology. Fascinated with the potential for machine learning and intelligence while also being filled with trepidation over it's possible...