Table of Contents
Getting Real about Real Time: 100G Virtual Networks Require Selective Real-Time Data Processing
Network and service support systems implemented prior to 2001 were not built to operate efficiently with data moving at Gigabits per second ( G). The majority of support systems active in networks today were placed into operation well before 2001. This is a problem that will only grow more challenging as network speeds and data volumes increase more quickly than the ability of communications service providers (CSPs) to manage them with currently installed systems.
Already, G is the norm in the network core; and DOCOMO PACIFIC is now beginning to build G networks, with the help of Ciena. Verizon explained that it is doing likewise, and has added Cisco into its partner mix. Granted, DOCOMO’s service will not be delivered on a single G pipe, but through a mix of G and G facilities. Still, any data collection system in this pathway will look at an equivalent G pipe filled with traffic.
Engineering plans for delivering G speeds are not far behind. Collecting and managing such large data volumes is challenging in traditional physical network environments. The challenge comes not only from growing volumes, but also from the inherent transmission errors and dropped packets that come with higher speeds. Adding Network Functions Virtualization (NFV), Internet of Things (IoT)-based services, time-sensitive smart cars and smart cities, only escalates the challenge. Packet delay is intolerable for these services, all of which require near real-time communication. Network probe providers do a commendable job keeping pace with evolving networks by developing the capacity to collect most everything thrown their way; but even network probe vendors need help at times, from accelerators, load balancers, and storage solutions.
In response to virtualization, the need arose for the ability to dynamically collect data when and where it is most needed. Also, the need for a lower-cost and a more flexible alternative to hardwarebased network probes drove the industry to a software-based probe model. However, as flexible as software can be, its inability to process the variety and volume of data packet types in real-time makes even more processing and data delivery help necessary.
In addition, because so many departments within the CSP business require access to network data, and view the importance of certain data sets as more germane to their goals, processing and delivering customized data sets is growing more complex.
Get Industry Insights. Simply.
Talk to Ahmad
+1 718 618 4302
"Connectivity comprises only 14% of mobile IoT and LPWA solutions revenue and it is unclear how operators will successfully generate value beyond this." Total revenue from the mobile IoT and LPWA solutions ...
“Growing need for centralized monitoring and predictive maintenance of manufacturing infrastructure is boosting the growth of the IoT in manufacturing market” The Internet of Things (IoT) in manufacturing ...
Use this report to: - Learn about the important role that sensors play in Internet of Things (IoT) applications. - Gain an insight in to the opportunities that the IoT provides for sensor manufacturers. ...