Tuesday, March 3, 2015

Blueprint: Why the IoT is Forcing Mobile Operators to Redesign

by Frank Yue, Senior Technical Marketing Manager, F5 Networks

The Internet of Things (IoT) is composed of automated machine-to-machine (M2M) devices and is poised to be a huge boon for consumers everywhere. People and businesses will be able to access data from wearable exercise monitors, home automation and security—and pretty much any device imaginable. Manufacturers are connecting water meters, cars, and even crockpots to the Internet. Some estimates put the number of IoT devices at 16 billion in 2014, growing to over 40 billion by 2020. By comparison, there are predicted to be 8.3 billion non-M2M devices by 2019.

This surge in the number of automated devices connected to the Internet will significantly impact service provider networks. M2M and automated products behave very differently when compared to biologically driven devices. Mobile service providers must take a close look at how these devices function and how they interact with the Internet through their data connections as they continue to build and update their advanced LTE mobile networks.

Until now, consumer driven connections through mobile phones, tablets, and PCs have been the focus for the design and build of LTE networks. These devices and connections consume large amounts of data as everyone's hunger for more mobile content is sated. Although generally random, the connections are statistically predictable. Service providers are using this predictability to ensure that the networks they build can support the planned demand.

Not all connections are created equal

M2M devices have an extremely different connection profile from non-M2M devices. They are designed to send periodic updates with their current state and other data. Although the amount of content is small, there is a specific cadence to the process of delivering these updates. Currently, most service providers’ networks are not designed to handle this regular connectivity.

In the 3GPP LTE network architecture, there are approximately 26(!) steps for any user equipment (UE)—including IoT devices—to attach (register) to a mobile network. When a device is idle for a period of time, it goes to sleep and detaches itself from the network. The idle timer is usually fairly short since the device is tying up precious radio bandwidth as long as it is attached (registered). In order to re-attach, the device must go through those 26 steps again.

A typical M2M device will send updates on a periodic interval such as every minute, five minutes, or 15 minutes. In addition, these devices will often send more comprehensive updates on a larger interval, often hourly or daily. For instance, a wearable training device may send updates to the Internet every 5 minutes while a residential water meter will only send a daily message to update the home’s water consumption.

All of these updates means there will inevitably be surges of registrations and floods of small chunks of data on the mobile networks. The minute-by-minute updates may lead to only small surges, but the more comprehensive updates (such as hourly or daily) may cause far larger surges. So, the raw volume of data is not the main issue; it is the number of connections and reconnections and the number of devices that is the primary concern.

The mobile service provider is not ready to handle these “harmonic surges” in registrations and updates. When a device wants to send more data, it will need to register to the network again. These surges in registration messages can easily overload the Diameter signaling infrastructure.

Cloud technologies are not just for the cloud

To properly and efficiently handle the connection and data patterns that IoT creates, mobile service providers need networks that are agile and elastic in order to create on-demand infrastructures that can support these surges. At the same time, they want to avoid the expense of purchasing and maintaining equipment that is unused except during these surge periods.
This is where service providers can leverage virtualized and cloud technologies. Network Functions Virtualization (NFV) is the architecture that can transform today’s physical networks into flexible and dynamic infrastructures that run services in cloud environments. Cloud technologies are designed to deliver the on-demand resourcing, or elasticity, that is necessary to support traffic surges. This maps directly to the planned capabilities of NFV.
An unspoken goal of NFV is to transform the mobile service provider’s core network into a cloud-like environment with all the business and operational benefits of an agile and virtualized infrastructure. To support the demands created by the growth of IoT, the registration and policy infrastructure of the evolved packet core (EPC) must be virtualized.
Note, however, that virtualization, in and of itself, does not deliver the required elasticity. On-demand resourcing is only possible with the abstraction, programmability, and orchestration of these virtualized services. The detection of surge periods must be coordinated with the spinning up of the resources required to meet the expected demand. After the flood of connections pass, these resources must also be freed up to be available for other purposes.

“Harmony,” not “harmonics”

The potential for the IoT to overwhelm the mobile network infrastructure is very real. And with the number and types of connected devices growing exponentially, the hype for IoT continues unabated.  

When it comes to building their infrastructures, mobile service providers need to work closely with the key IoT standards bodies and manufacturers to find common ground. If they do not address this growing type of connectivity, service providers will experience a devastatingly similar situation to the harmonic convergence of wind speed that caused the Tacoma Narrows Bridge to collapse in 1940

About the Author

Frank Yue is the Senior Technical Marketing Manager for the Service Provider business at F5 Networks. In this role, Yue is responsible for evangelizing F5’s technologies and products before they come to market.

Prior to joining F5, Yue was sales engineer at BreakingPoint Systems, selling application aware traffic and security simulation solutions for the service provider market. Yue also worked at Cloudshield Technologies supporting customized DPI solutions, and at Foundry Networks as a global overlay for the ServerIron application delivery controller and traffic management product line. Yue has a degree in Biology from the University of Pennsylvania.

About F5 

 F5 (NASDAQ: FFIV) provides solutions for an application world. F5 helps organizations seamlessly scale cloud, data center, telecommunications, and software defined networking (SDN) deployments to successfully deliver applications and services to anyone, anywhere, at any time. F5 solutions broaden the reach of IT through an open, extensible framework and a rich partner ecosystem of leading technology and orchestration vendors. This approach lets customers pursue the infrastructure model that best fits their needs over time. The world’s largest businesses, service providers, government entities, and consumer brands rely on F5 to stay ahead of cloud, security, and mobility trends. For more information, go to f5.com.


Got an idea for a Blueprint column?  We welcome your ideas on next gen network architecture.
See our guidelines.