This white paper examines current service assurance systems and how they relate to Big Data, Artificial Intelligence, Machine Learning, and Deep Learning. Contrarily, the concept of Small Data – using active testing and monitoring to provide direct answers to the relevant service quality questions, is also introduced. Additionally included, are highlighted results of the NFV Service Assurance and Analytics research study and survey completed by Heavy Reading.
Key takeaways from the research include:
- The industry’s expectations on Big Data and AI should be lowered to a more realistic level. These technologies will not provide a panacea for all service assurance needs and transformation challenges.
- As Big Data and AI rely on relevant and high-quality data, large amounts of low-level data will not satisfy the requirements to receive satisfactory answers to the relevant service assurance questions. With current systems, it is very difficult to obtain high quality service-related data on network services using traditional infrastructure-centric assurance tools or Big Data and other AI technologies.
- Data from active testing and monitoring can provide detailed, real-time service KPIs. These KPIs, which can be referred to as Small Data, provide great value by themselves, but they are also enablers for the successful application of Big Data and AI.
Download the white paper to learn more.
Figure: AI, Machine Learning, and Deep Learning. Adapted from Nvidia blog (5).
Note: the illustrations in this diagram, they depict some examples of AI applications: game playing, spam filters, and catching cats on the Internet, which are revisited in the white paper.
By contrast, Small Data, obtained through active testing and monitoring, presents you directly with a pertinent handful of needles, bypassing the Big Data requirements as well as resource-hungry data lakes.