Current Filter: Network>>>>>Opinion>

   

The explosion of electronic information, coupled with the proliferation of increasingly sophisticated analytics, has transformed raw data into a profoundly powerful tool for insight and action. New Big Data use cases unfold as organisations look to harness, analyse, and correlate data sets to streamline operations, improve efficiency, connect with customers and increase revenue.

One such data set is network data. The maturation of Big Data and analytics technologies is enabling businesses to capture this data, not only for visibility into IT operations, but also for business analytics.

TRANSFORM NETWORK DATA TO WIRE DATA FOR ACTIONABLE INSIGHT
Raw network data - packets - is far too voluminous to be captured and interpreted on its own. Analysis of network traffic at 10 Gbps would produce more than 100 terabytes of raw data daily. Moreover, raw packets have no notion of application or session boundaries, making it nearly impossible to extract any meaningful or timely information. To make network data usable and scalable it must be transformed into wire data, and this requires three key elements.

First raw packets must be reassembled in real-time into a per-client session or full transaction stream based on its protocol. Second, the data must be decoded from their wire protocol streams with an understanding of each wire and application protocols' boundaries. Third, only meaningful results based analysis is extracted from the stream, indexed, and stored as metadata. This wire data set when analysed in real-time provides deep, granular insight into IT and business performance.

WIRE DATA: EVERYTHING TRANSACTS ON THE WIRE
Wire data is emerging as the richest source of observational insight for IT. It's like Google Earth for your IT environment and creates a new and rich source of insight regarding to whom machines, clients, users or applications are talking, what they are saying (the data payload) and how frequently, the volume of data exchanged, the transaction performance and their availability, to name just a few. This level of insight empowers IT teams to rapidly pinpoint and resolve issues, auto-map relationships between systems, understand anomalous behaviour, identify performance issues and optimisations, ensure compliance and drive operational efficiency.

This same data can also provide real, actionable business insights. Because everything transacts on the wire, the ability to extract, analyse, and visualise all transactions and any element of their payload can reveal real-time information about product sales, revenue, customer engagement, records access, patient wait times - the list goes on.

Here are just a few examples of how businesses can leverage network data that has been transformed into structured wire data:

• Correlate merchant revenue with payment transaction performance
• Track purchasing patterns by merchant, geography, user, etc.
• Monitor emergency room wait times decoding the HL7 protocol
• Real-time breach detection, including data exfiltration and unauthorised access
• PCI compliance by proving that all transactions in flight were encrypted.

MULTI-SOURCE CROSS-CORRELATING WIRE DATA: DEEPER, RICHER INSIGHT
When it comes to Big Data, the insights you derive depend on what you put in. As they say, garbage in, garbage out. Just as organisations don't expect to gain a complete understanding of purchasing behaviour by looking at order activity alone, they shouldn't expect one single data set to provide complete visibility across the entire IT environment.

With the correct data sets and Big Data architecture in place, IT can go from being a support centre to a core driver of operational and business insight. Networks and applications produce a wealth of data - including wire data, machine data, agent data, and probe data - which, if analysed and correlated, can help transform both IT and business operations.

The network contains a vast wealth of information, and we are only just beginning to realise the potential of that data to deliver tremendous business value and advantage. NC