The more intelligent a network, the better it’s able to protect, analyse and ultimately optimise. The concept of an intelligent network has long been the stuff of dreams – networks able to dynamically respond to threats, based on previous experiences – and of a few nightmares: the Skynets of fiction.
Over the last 20 years, we’ve slowly nurtured those visions and delivered them into our reality, led by solutions such as Scrutinizer by Plixer, which, since 2000, has been at the forefront of collecting, visualising, and reporting on every conversation that crosses networks, and the metadata generated therein. Amongst its competition, Scrutinizer is unique is that it does not require the implementation of agents; it collects content-rich metadata directly from existing network infrastructure (switches, routers, firewalls, etc.). It, in other words, offers more – more insight, control, intelligence – without requiring additional overhead or resources.
Network Intelligence module: a step beyond analysis
The challenge of data is quantity. Depending on the network size, thousands or millions of connections will be reported on and visualised at any time. The ability to detect patterns within these data streams is critical, which first requires different data streams to be collated and analysed within the same platform. Maximum insight is derived from maximising oversight, and Scrutinizer by Plixer excels at combining data from all network connections, and offering a single reference point.
Plixer’s new Network Intelligence module, however, takes this a step further. It uses machine learning (ML) to augment NetOps staff capabilities, and applies ML algorithms that are specifically tuned to network assets (such as switches and routers) rather than end devices.
Real-time metadata collected by Scrutinizer is streamed to the module, which uses it to dynamically monitor, baseline and predict network and WAN utilisation. Foremost, it helps to optimise the network performance by identifying, in real-time, bottlenecks and peaks, and proactively calculates capacity requirements before service degradation can occur.
Proactive versus reactive
Our world is increasingly becoming one that necessitates proactivity. It’s no longer enough to react – no matter how timely – to network performance drops or security issues. Even five minutes of degradation is enough to begin a cascade of consequence. Now, through Scrutinizer, network managers have access to a ready supply of historical data, which is usable by both NetOps and SecOps teams (we spoke about the benefits of having a single source of data for both teams in a recent blog post).
Network Intelligence helps protect against losses. By applying ML to every network conversation, it is able to provide advance notice of capacity requirements, and delivers the empirical data needed to obtain C-Level approvals before users are impacted. Network managers are consequently able to get in front of problems, and proactively plan, manoeuvre and resolve.
Lastly, for organisations with centralised data lakes, in which organisation data is consolidated, Network Intelligence enables the streaming of network traffic intelligence using Kafka, in real-time. Network managers are able to configure what data is streamed, and how it is aggregated, to ensure the relevant data is available, thereby maximising insight and value. Through this, Scrutinizer helps increase the ROI of existing network infrastructure, making more of what already exists.