Powerful smartphones can do a lot, but they also carry a number of security risks. Claus Eßmann and Dr. Peter Krüssel thought about the alternative: the cloudphone with an operating system from the cloud and short latency times thanks to edge computing.
Smartphones today are actually "mini-computers" in your pocket. We can run all kinds of apps with them - and make phone calls on the side. The performance of these pocket computers has not yet reached its limit: The top models now have SDKs for the processors used to create local artificial intelligence applications. In addition, these processors are now increasingly found in large server farms that are committed to energy efficiency. At the same time, smartphones now have more main memory than the average PC did a few years ago.
More power, more security hazards
However, the security threats for smartphones are also growing with their performance: Viruses and Trojans are now available for Android smartphones in almost the same number as they are for Microsoft Windows-based PCs. Repeatedly security gaps are discovered and the manufacturers can hardly keep up with the delivery of security patches. When the security patch is delivered, only 30 percent of smartphone users install it. What sounds like a merely pessimistic scenario for private users is a real horror scenario for business users and corporate IT administrators.
But is this scenario inevitable? How far can one go with existing and future technologies without limiting the user too much? Today's solutions mostly rely on mobile device management. Here, certain parameters of the smartphone are administered centrally. This creates a certain level of security, but the user still has to take care of software updates and other security measures himself.
Easy and secure via the cloud?
What if you basically only used the devices to visualize content and ran the operating software, for example Android or iOS or another ecosystem, from the cloud? That would mean that the smartphone would no longer have to be so smart, since no more calculations would be performed on the device, but only in the cloud. The results of the calculations are sent to the device as graphics for visualization and interaction with the user. The software and the user data are centrally managed and stored in the cloud, so you can switch to another device at any time without losing data or a workflow! The devices used only have to take care of the visualization and interaction with the user and therefore require less computing power and memory. This in turn could increase the runtime of the devices to a duration never before reached for smartphones.
In addition to the outlined security aspect, a possible central update, feature and patch management, the central and non-local data storage as well as the higher battery life of these cloudphones, another essential advantage can be listed: In all probability, the end devices would be significantly cheaper to manufacture than smartphones - with the corresponding positive consequences for users and the possibility of a "smartphone as a service" business model for network operators.
The “Cloudphones” concept requires full network coverage
However, in order to make this concept of a "cloudphone", whose intelligence lies in the cloud a reality, a number of preconditions must be created. High data transfer rates, low latencies and almost 100% network coverage - both indoors and outdoors - are required.
Network coverage, both national and international, is important, because if the concept is implemented consistently, which does not provide for the execution of any local programs on the device, the device without an Internet connection will no longer have any function. At the same time, a relatively high bandwidth is required, since - similar to the "cloud gaming" that is currently becoming fashionable - a lot of data has to be sent between the device and the cloud, and the device will not be accepted by the user if the content builds up slowly on the screen and the operation of the device is therefore perceived as "tough". However, bandwidth is only one aspect that influences the user experience when operating the device. If the latency is too high, the user has the feeling that nothing happens for a long time after an interaction, i.e. for the user, the reaction of the device, although coming from the cloud, should take place in quasi real time. The customer experience would thus have to be on approximately the same level as with previous smartphones.
While LTE can certainly provide the necessary bandwidth, the general latency of 50 to 80 ms is no longer sufficient for this fast response. What now? Can this idea only be realized when 5G is introduced across the board in a few years' time?
Realization with Edge Computing
No, one answer to the question of low latency can be edge computing. Here, the response times are drastically reduced by moving the cloud (or rather, part of the cloud) in which the calculations take place closer to the device and the user. Instead of distances of a few hundred kilometers between the device and the cloud, Edge Computing aims for a distance of less than a hundred kilometers, which significantly shortens the signal propagation time.
In addition to ensuring such a solution with the appropriate network qualities, functionalities similar to those of the established OTT providers must of course also be offered. In contrast to the OTTs' systems, such as Apple, which are virtually closed in the network due to the combination of terminal device, operating system and apps worlds, these cloudphone-based offerings from the carriers would have to have an open character and offer a comparable spectrum of services, applications, functionalities and user experience. This does not necessarily mean that telcos have to operate completely new app stores. Rather, there are various possibilities for reusing existing resources: On the one hand, carriers can access apps from the FirefoxOS and ChromeOS app stores, and on the other hand, the virtualization of complete apps, such as Android apps, offers the possibility of integrating existing apps into a new, virtual app store.
Courage for a new business model
In conclusion, it can be said that with this approach, operators could consistently exploit their core asset, the control over the telecommunications network, to the benefit of their own customers end-to-end. A prerequisite is quality-assured treatment of customers in the network who rely on these virtualized end devices and solutions. Carriers could use the cloudphone approach as a means of counteracting the bargaining power of the major end device manufacturers. They could also enter the established business models of OTTs with their closed user groups and fully integrated eco-systems. There would be an opportunity to strengthen their own customer relationships, at least in certain segments, and to generate new sources of revenue. However, this would mean a radical departure from existing business models.
