Blog

22nd March 2016

Enable Proactive Service Assurance with Predictive Failure Identification

The Value of Fault Management Analytics, Part II

Proactive Service Assurance
In today’s highly competitive market, it’s challenging to meet expectations for quickly resolving network issues, minimizing fault discovery times and preventing outages with the increase in network demands and limits on budget and head-count.

Implementing smarter infrastructure with advanced analytical algorithms and techniques provides the ability to evaluate more information and automatically get actionable insights. This insight enables Network and Services Operations Center (NSOC) staff to be more proactive in the management of evolving networks and prevent network outages.

Predictive Alarms
Network operations controllers are flooded with hundreds of alarms every minute, to the point that they become overwhelmed. NSOC staff rely on experience to identify and focus on the most critical ones. But advance warning of an upcoming malfunction may be hidden in the alarms and go undetected until it becomes too late. Critical warnings and early signs of problems are too easily overlooked.

With most trend reports, NSOC personnel must identify and analyze trends based on attributes they believe may be of interest, creating a subset of defined trends, which means that some potential trends and problems are left out of the reports entirely. And due to the dynamic nature of todays’ networks, these defined trends need to be constantly updated to prevent a malfunction.

TEOCO’s Predictive Failure Identification analytical algorithms help NOC professionals by constantly monitoring the large number of alarms along with their attributes and history. Alerts are sent for outliers and behavior trends that have a high probability of predicting malfunctions. These alerts then can be managed with corrective action so malfunctions and outages are prevented in advance.

Predictive Risk Rank
Most large networks have hundreds or even thousands of network elements from different vendors with different “ages”, utilization and operational patterns. Some of these elements are more prone to errors and faults than others which makes them easy to detect in the usual course of operations. But certain elements prone to errors and faults are not as easy to detect, simply because the faulty behavior is not “dramatic” enough. This is the case with slow performance degradation, erratic behavior of errors or small changes in the rate of errors. These might not trigger an obvious major or critical alarm, yet they can be detected using advanced mathematical algorithms. By looking at multiple parameters that signal a high risk element, they can be identified and ranked as high risk.

TEOCO’s Predictive Failure Identification lists elements by instances, Network Element type, geography and more, and it allows the NOC personnel to start investigating the high-risk objects. The combination of the listed high-risk objects with their behavior trends provides the information needed to investigate and solve problems quickly.

Preventing Network Outages
Predictive Failure Identification is part of TEOCO’s Helix Unified Service Assurance Software Suite, detects deep trends and predicts failures in advance, to equip staff with the necessary tools to successfully prevent network malfunctions and minimize service outages.

The Value of Fault Management Analytics
TEOCO’s Fault Management Analytics delivers the power of advanced analytical algorithms to both the NOC and the engineering teams. This powerful analytical functionality is integrated across TEOCO’s Service Assurance solutions and provides added benefits, especially when used with its Performance Management and Customer Experience Management modules.

graph34

If you missed it, click here to read Part I of The Value of Fault Management Analytics to learn how to turbo charge service assurance with Automatic Route-Cause Analysis.

For more information on Helix, TEOCO’s Fault Management Solution, click here.

Sign Up for Industry Insights and News

You have Successfully Subscribed!