Managed Systems Monitoring

Because Photon and the other services in Breakpoint’s Vantage suite share a common technology core for end-to-end service monitoring, it offers a single pane of glass view of your service landscape with monitoring using synthetics, analytics, real user monitoring and systems monitoring in an integrated manner. It is easy to create a dashboard that correlates the user experience with any available systems metric or complex business analytics.

Photon is designed and built with robust mechanisms to deal with network interruptions or service disruptions, so that you don’t lose metrics. When the agent is not able to hand off metrics to the data store, it will buffer the metrics up to a configurable limit. An additional caching mechanism on the VantageD collector, which is deployed in close proximity to the monitored nodes, ensures highly reliable metric delivery for your critical or time sensitive applications.

Most of the leading monitoring solutions provide at best one minute resolution, which is insufficient resolution for troubleshooting highly volatile or transient application or infrastructure issues. Photon is able to provide up to 10 second resolution for nearly all available metrics, along with reliable ingestion and visualisation of high velocity monitoring data.

Collecting and storing metrics are only half of the solution. One has to present these metrics in a meaningful and easy to interpret manner in order to make the data useful. Part of our service is for our experts to build out the real-time dashboards required to visualise monitoring data for various stakeholders according to their needs, and to maintain these dashboards on an ongoing basis with new or changing requirements.

Real time, operational needs are different from reporting needs, which focus more on long term trends, forecasts and capacity planning elements. Our systems expertise, combined with data analysis and visualisation skills, means we know what data to look at in order to inform capacity planning and forecasting needs.

Monitoring environments are complex and data rich, and as such they require careful management and maintenance in order to perform reliably and optimally. Moreover, monitored environments are almost never static. This part of our service ensures that the databases, software stacks, systems and agents are optimally configured, maintained and updated. We also add and remove nodes that are introduced or decommissioned. This good hygiene and housekeeping ensures continued optimal output from your monitoring investment.

Standard, out of the box monitoring is almost never sufficient when it comes to optimising application performance, or troubleshooting performance or reliability issues. However, building out visibility and insights into the more esoteric aspects of app or infrastructure issues can be incredibly difficult. Our experienced team is able to work with our customers and their application vendors to provide instrumentation and hooks into applications, or read from any exposed interface or log file, in order to extract and visualise the necessary data for monitoring purposes.

A common failure of monitoring deployments relate to improper implementation of alert thresholds – not getting alerts for critical conditions means that impactful conditions get overlooked. Even worse, too many alerts could lead to fatigue and a low signal-to-noise ratio, which means that serious problems get lost in the noise. Our team is dedicated to having the finger on the pulse, and are constantly tuning thresholds based on observed baselines and system behaviour, informed by service quality metrics.