Driven by the explosion of data, the core architecture of the Internet of Thing, broadband networks, and next generation cellular, are all about to be dramatically impacted by the federation of data over the topology.  Cloudlets are emerging as the next, most import IT innovation in the future decade.  Are you ready for cloudlets?

So, what is a cloudlet?

According to Wikipedia, A cloudlet is a mobility-enhanced small-scale cloud data centre that is located at the edge of the Internet.  The main purpose of the cloudlet is supporting resource-intensive and interactive mobile applications by providing powerful computing resources to mobile devices with lower latency.

Cloudlet 2

In the beginning of the concept, way back in 2016, this definition would have been correct.  However, it is already expanding and diversifying far beyond this initial description.

Depending upon who you are talking too about cloudlets, they have several different names.  They can be known as Fog Computing and Edge Computing too.

The cloudlet term is driven primarily by the ISO / IEC JTC-1 SC38 standardization subcommittee working group. This is the Cloud group extending its dominion out onto the network fabric all the way to the edge.

Fog Computing is term from Cisco, so many standards bodies are avoiding it as it is vendor specific.  Edge Computing is also coming from the standards setting communities to avoid the vendor linkage.  At ISO / IEC JTC-1 SC-41, is a network centric standardization subcommittee working group that is network focused for Edge Computing.  So, Sc-41 sees these nodes as a component of the network, not a component of the cloud like SC-38.

The cloud folks see cloudlets as a direct extension from the cloud.  Whereas, the network folks see Fog and Edge as centric to the network itself, and simply linked to the cloud.  It is this centricity and focus that creates debate and conflict in the IT world.

Cloudlet 5

In reality, all of these points of view hold merit.  It all depends upon how you deploy this solution.

Cloudlets will begin to increase in visibility and importance in 2019 and beyond as they are a vital component of the emerging 5G cellular design.  IoT uses them already to push intelligence to the edge of the network and bring important performance characteristics to the edge of the network.

By pushing the intelligence towards the outer edge of the network, we realize several important features:

  • Reduced latency as the sensor and its related compute device are much closer together, so decisions are made in nanoseconds and not milliseconds or even seconds
  • Less traffic on the network fabric because not all traffic needs to traverse the network to get to the cloud.  Traffic that is exception-based can flow to the cloud, but status quo traffic does not really need move upstream, just a “keep alive” pulse to tell the cloud that the edge device is active and function normally, or derived summary data datagram that summarizes the edge activity is all that needs to travel to the centre of the network
  • Compute can now flex to be associated with travelling nomadic data from mobile devices that come and go at different locations
  • Often, the network has insufficient resources to carry the mass volume of data flows, so by offloading data, and keeping its residency to the network’s edge, we can reduce the burden on the network fabric and avoid costly upgrades
  • Security will be impacted and can be enhanced through segmentation and isolation.  However, it could also be a risk if not addressed correctly, so it might be a point of entry for intruders.  Therefore, security must dynamically map to this new architecture.  The old centralized designs are not going to work effectively with this federated model

Many IoT experts, including me, believe that cloudlets will emerge as a class of computing unto itself, in a line that started with desktops, then mobile apps, then data centers, and large-scale clouds. The apps that emerged in each type of computing class were only capable with each new class — none could have existed in the previous era alone.  The same will be true with cloudlets whereby new apps designed just to function at the edge, or in coordination with the centre will be developed and deployed.

Cloudlet 1

The edge network technology will have a latency of less than 50 milliseconds. This ultra low latency will drive another attribute of the Internet of Things and 5G connections – the way that the end node ‘couples’ to the network, cloud, and end user devices.  The ecosystem coupling will change with the advent of cloudlets.  The coupling, which has historically been upstream towards the centre will continue, but will be loosely coupled or sporadically uncoupled.  It is predicted that the cloudlet coupling will be focused much more on the end user.  It will be headed downstream and will be tightly coupled towards the end node devices adding small site compute, storage, and analytics to enhance the richness of the experience.  Node to node lateral connections (peer-to-peer) will further fuel the effectiveness and connectedness for the end users.  Why send the data all the way to the core cloud when it can be determined adjacent to the serving node by other local nodes.  So, data may still flow upstream, but not in an urgent manner to best service the needs of the end user.  It can flow as and when needed.  Likewise, external data or archived data may flow downstream from the core cloud to the serving node, like regional weather data.  Again, it is not needed urgently so it can be updated in a more relaxed time frame compared to normal interactions by the end users.

The weight of the computational intensity will dictate how the end user connects and where they connect to.  Heavily weighted data applications, such as virtual reality will continue to be vector math based monsters that depend a bare metal solutions that can only be found in the cloud.  However, lighter weight applications, such as local weather readings coupled with regional and national readings can live at the edge in the cloudlet and collaborate with the cloud whenever and as necessary.

Cloudlet 6

New tiers of lightweight computing will extend the edge computing nodes outwards closer to the end users.  Cisco calls it Mist Computing to extend their cloud / Fog / Mist analogy.  Mist compute clouds are typically self-organizing, in that they are instantiated on more of an ad-hoc manner by the individual element’s desire to interoperate or operate as a single entity. This is contrary to fog computing infrastructures, which follow the cloud or network centric deployment models, under the ultimate ownership of a public or private cloud service provider or a network operator depending upon the application.  End user applications would likely be cloud connected, but infrastructure focused applications, for security as an example, would be managed from the Network Operations Centre (NOC) or a Security Operations Centre (SOC).  Indeed, as of this post, the NIST has a draft definition in circulation outlining these variants, along with introducing Mist into its vocabulary.1 These fog nodes might be characterized by the fact that they have higher compute capabilities and higher bandwidth connections to the compute cloud.

Cloudlet 4

There is still a great deal more work to do on cloudlets and this architecture and network topology is still evolving and subject to change.  Now is the time to dream up the next generation of innovation and shift the paradigms in new directions.  Data will reside ubiquitously all over the network, so management of this data, retrieval, and archiving of it will take on new challenges that will need to be overcome.  There is much work to do on this topic before we can truly understand how it will affect our lives going forward.


Notes:

1./ Michael J Martin is a member of the NIST SP 500-325 Fog Computing Conceptual Model, National Institute of Standards and Technology. committee that wrote these definitions.  A link to the document will be provided once the USA resolves its current funding problems.  As a result, the NIST web site is currently not available.


References:

Dredge, S. (2018). 5G Multi-Access Edge Computing with cloudlets in fog creating mist, and why the hell does every networking journal read like a London weather report these days? metaswitch. Retrieved on December 28, 2018 from, https://www.metaswitch.com/blog/5g-multi-access-edge-computing-with-cloudlets-in-fog-creating-mist

LeFebvre, R. (2018). Carnegie Mellon’s clear view on 5G cloudlets. Superuser. Retrieved on December 28, 2018 from, http://superuser.openstack.org/articles/future-5g-cloudlets-already-carnegie-mellon-university/

Wikipedia. (2018). Cloudlets. Retrieved on December 28, 2018 from, https://en.wikipedia.org/wiki/Cloudlet

Zhang, L. (2018). Bringing IoT to the Cloud: Fog Computing and Cloudlets. IoT Zone. Retrieved on December 28, 2018 from, https://dzone.com/articles/bringing-iot-to-the-cloud-fog-computing-and-cloudl


About the Author:

Michael Martin has more than 35 years of experience in systems design for broadband networks, optical fibre, wireless and digital communications technologies.

He is a Senior Executive with IBM Canada’s Office of the CTO, Global Services. Over the past 14 years with IBM, he has worked in the GBS Global Center of Competency for Energy and Utilities and the GTS Global Center of Excellence for Energy and Utilities. He was previously a founding partner and President of MICAN Communications and before that was President of Comlink Systems Limited and Ensat Broadcast Services, Inc., both divisions of Cygnal Technologies Corporation (CYN: TSX).

Martin currently serves on the Board of Directors for TeraGo Inc (TGO: TSX) and previously served on the Board of Directors for Avante Logixx Inc. (XX: TSX.V). 

He serves as a Member, SCC ISO-IEC JTC 1/SC-41 – Internet of Things and related technologies, ISO – International Organization for Standardization, and as a member of the NIST SP 500-325 Fog Computing Conceptual Model, National Institute of Standards and Technology.

He served on the Board of Governors of the University of Ontario Institute of Technology (UOIT) and on the Board of Advisers of five different Colleges in Ontario.  For 16 years he served on the Board of the Society of Motion Picture and Television Engineers (SMPTE), Toronto Section. 

He holds three master’s degrees, in business (MBA), communication (MA), and education (MEd). As well, he has diplomas and certifications in business, computer programming, internetworking, project management, media, photography, and communication technology.