Blog Post

Rocana Ops 2.0: A Deep Dive into Rocana Reflex

    
January 30, 2017 Author: Eric Sammer

Today, users of Rocana Ops have access to all event and metric data across all systems, down to the finest granularity (yes, including debug level), over time, in real time. Rocana Reflex adds the ability to take action on that data, enabling a new set of use cases – from simple alerting, to advanced automation and data-driven decision making.

reflex-page.png

Leveraging the fact that Rocana Ops is a first class data warehouse for event-oriented data (rather than just another siloed monitoring tool), Reflex is a full-featured complex event processing (CEP) system that complements the data integration, massively parallel storage and query, data management and governance, and analytics features present in the system today. This means Reflex operates directly on the data as it streams into Rocana Ops, in real time, with the ability to transform, aggregate, and otherwise munge both events and metrics over tumbling and sliding windows. Actions can be attached to Reflex triggers which allow you to send emails or make REST API calls in response to specific conditions allowing you to integrate with case management, notification and escalation systems, and even internal tools and systems.

As a simple but powerful example of Reflex and other Rocana Ops capabilities in action, consider a Pivotal Cloud Foundry (PCF) use case where the goal is to instantly and automatically open a ServiceNow ticket any time a PCF application crashes. The following demo video shows how easy it is to do this with Rocana Ops 2.0.

Furthermore, Reflex triggers are decoupled from actions, meaning that rather than simply invoking an action when a trigger fires, an event is produced within Rocana Ops. While one or more actions are typically fired in response to this event - and we make that easy to configure - these events can be reported on, searched, graphed, or analyzed just as any other event in the system. In fact, other Reflex rules can take these events into account when looking for even more complex patterns. Since a trigger firing is just an event, it’s even possible for Rocana’s anomaly detection to detect abnormal behavior in the rate of a particular trigger firing, for example.

Reflex is different than the basic alerting functionality in most monitoring products. It operates across all your infrastructure at a scale that most of those systems can not: 10s to 100s of terabytes per day. Rather than polling queries and saved searches, Reflex is a streaming system operating in real time. It understands more than just metric data with thresholds, opening up a world of smarter triggers and fewer false positives or false fires. Since triggers produce events, and events are just messages in Kafka in an open source data format, they’re useful beyond the functions of Rocana Ops itself. This means your own applications can instantly detect and respond to complex patterns in event data without being in the critical path of alerts or automation tasks.

Reflex is part of Rocana Ops. There’s no additional licensing or cost. That’s a big deal.

You and your team can and should have nice things. While organizations stack up resources to capture, process, store, query, analyze, and act on “business data,” ops teams rarely see the fruits of that investment. Rocana Ops 2.0 brings all of the advanced functionality of a full-featured data warehouse to those that make all of those systems run.

Interested in finding out more? Be on the lookout for an upcoming blog post in which we’ll be taking a deeper dive into the all-new first responder experience in Rocana Ops 2.0!


Learn More...