Is a Streaming Analytics Correlation Engine powering your big data?

Hype or no hype, big data is not a new phenomenon and can provide businesses incredible insight. That is, if businesses know what to look for, how to use the data and have the right tools to hand. In this blog post we use insight collected from our telco operator customers to demonstrate real value from big data.

Knowing what you’ll know tomorrow, today

You’re collecting performance data from your Radio Network at 15 minute intervals, resulting in hundreds of millions of performance counters being loaded to your Big Data Analytics database, each and every day. KPIs are calculated and published in reports which are then scrutinised to determine what problems occurred on the network… Yesterday, not now.

Sound familiar?

As a telco operator you have 30,000 3G cells in your network and you want to monitor 20 KPIs across all cells e.g. DCR, CSSR, etc. Surely, you also want to know if any cell or group of cells (e.g. town, city, region) breaches a defined threshold for each KPI as soon as it happens. Therefore, if you assume each KPI formula uses four raw performance counters, your correlation engine needs to receive and process 2.4 million relevant counters per 15 minute performance management (PM) data collection period whilst maintaining 600,000 discreet monitoring scenarios simultaneously and continuously.

Power in the engine

A Streaming Analytics Correlation Engine, such as ZenPM™, scales to give you the power to monitor the necessary volumes of KPIs in real-time. It is purpose built to handle the massive scope and volumes that the big data revolution dictates. Thus giving you visibility of your network in real-time, enabling you to take swift action if a problem occurs…. Now, not tomorrow.

Share your experiences of finding value in big data with your peers by leaving a comment below.