Big Data: A Closer Look with Guavus

The way the network world has been thinking will change. OSS and element manager players—companies whose products are tied to a specific element—will go away, and a new breed of analytical applications will rise. Optimization will extend beyond network and traffic and down to application and subscriber.

Software-level solutions are crucial. For VoLTE to work, for example, CSPs need to follow the service across multiple segments of the network, and this will be ever more complex. And frankly, as these cases become more complex, and there is greater and greater service proliferation, the more Guavus' analytics applications are at the center of the intelligence ecosystem.

What advice would you give to CSPs regarding their big data analytics strategy?

In telecom and enterprise, the approach we recommend is to leverage existing infrastructure; don’t replace what has been done. Instead, add and augment with real-time streaming data and operational intelligence. Since many of the initial data implementations are focused on batch, offline data, it makes sense to add real-time gradually, as opposed to a full replacement solution. It's the combination of the two, stream and stored data, that will provide a fuller picture of what is happening in your network, when and ultimately how to take action.