Cyber Security

Part 1 : Edge Computing IoT & the scalability of virtual infrastructure

The Edge is starting to disappear from view, slowly being covered by the Fog – metaphorically at least! Whilst Edge computing has become more commonplace with more people interacting with Edge computing every day, it is not always fully understood. Edge computing, in a nutshell, is the concept in which computing is being brought as close to the source of data as possible, instead of centralising it at a data-processing warehouse. This allows for IoT devices to carry out basic processing tasks themselves or a local computer/server, reducing data output before it is sent to the Cloud.

Written by

Team Nucleus

Content
Written on

6th March, 2020

SHARE ARTICLE

The Edge is starting to disappear from view, slowly being covered by the Fog – metaphorically at least!


Whilst Edge computing has become more commonplace with more people interacting with Edge computing every day, it is not always fully understood. Edge computing, in a nutshell, is the concept in which computing is being brought as close to the source of data as possible, instead of centralising it at a data-processing warehouse. This allows for IoT devices to carry out basic processing tasks themselves or a local computer/server, reducing data output before it is sent to the Cloud.


Reducing the overall processing power of the Cloud server results in reduced bandwidth usage and reduced latency (as data does not need to travel as far). More importantly, this will increase the Quality of Experience (QoE) for the end-user. This is especially important for new technologies such as self-driving cars and virtual reality gaming.


As we start to approach the Internet of Everything (IoE) our smart cameras, printers, toasters and even fridge/ freezers are now connecting to the internet. A report by Statista suggests that by 2025 there will be in excess of 75 billion IoT devices worldwide.


More devices connecting to the IoT drives technological progressions such as advanced analytics and machine learning, which inevitably requires more resources. The computational power of Edge devices may no longer be sufficient to handle certain processes pushing data back to the Cloud. But what happens when the capability or the demand for the Cloud is too excessive for the task at hand?



Fogging (aka Fog Computing or Fog Networking) looks to seize this opportunity creating a more advanced and technically capable device which can complete advanced analytics and machine-learning processes that Edge computing cannot handle. The Fog also allows for mitigation against latency by preventing reaching back to the Cloud. Fogging is not a replacement technology; it is a complementary technology that supports the workload balance between the Cloud and the Edge to accommodate the evolution of the IoT.


The importance of this technology is reinforced by current industry buy in with the likes of Intel, nVidia and NGD Systems investing heavily in AI accelerator technology, which offers significant support in Edge and Fog computing.


NGD Systems have recently announced the availability of the first scalable, Edge ready M.2 form factor storage solution, delivering up to 8TB with AI-ready computational storage. This offers Edge and Fog infrastructure unprecedented scalability and capacity with minimal footprint, providing a great amount of opportunity to Open Infrastructure and hyperscale environments for both providers and consumers.


With the increase in Edge/ Fog technology, there will highly likely be an increase in threat actors trying to exploit these capabilities for their own nefarious reasons. For example, the creation of large scale botnets such as the Mirai botnet, which accumulated 2.5 million bots at its peak. Whilst this seems large, it may be relatively small compared to the size and processing power of botnets of the future looking to exploit these capabilities.

NUCLEUS

Recommended Posts

Subscribe to Nucleus blog updates.

Subscribe to our newsletter and stay updated.

Subscribe to Nucleus