Adoption trends in some mobile edge computing markets are clear, but questions remain
In a recent Mobile Edge Forum virtual event session (available on demand), Tantra Analyst Founder and Principal Prakash Sangam said that stakeholders should be encouraged to invest in mobile edge computing. We shared our thoughts on what to consider and how businesses and consumers can benefit from decentralization. cloud computing infrastructure. Sangam presented his four vectors and comprehensive capabilities that determine the success of mobile edge computing.
Sangam’s four vectors shaping mobile edge computing are reliable connectivity, ultra-low latency, extreme capacity, and power-efficient computing for artificial intelligence and machine learning. Readers of RCR Wireless News will no doubt find these first three vectors to be key elements of 5G’s value proposition, but semiconductor companies will need advances in chip design and the latest generation of cellular air interfaces. The former is definitely maturing as it seeks to power new experiences.
Regarding latency, Sangam pointed out that this is not just an application latency, computing latency, or latency on the wireless link. Rather, it is a measure of the total latency required to deliver a particular result. As he puts it, “end-user-experienced latency.” For network operators, compute providers, device OEMs, and other parties, it delivers the performance thresholds for quality of service and quality of experience, and simply delivers the low latency required for enterprise compliance. to make sure it provides accurate delays without delay. Service Level Agreement.
Seeing the need for power-efficient computing, Sangam notes that edge computing nodes by their very nature offer margins in terms of space, power, heat dissipation, etc. compared to centralized cloud computing facilities. pointed out that there is no “When you look at edge cloud deployments, they are very constrained,” he said. “Considering all this, computing has to be very power efficient…AI/ML in the edge cloud is a big topic in its own right.”
He continues: But it’s the ability to extend across these vectors, not just provide them all… if you have to peak all these vectors implement and expand what you might not cost-effectiveness of the applications you are running… The overall idea is to scale your network across these vectors to reduce the cost required for a particular type of application or service. It’s about providing high-performance connectivity and computing. That’s the key to providing superior superiority. ”
Questions to ponder
With the obvious opportunities for mobile edge computing (for example, Gartner’s prediction that more than half of enterprise-managed data will be created and processed outside the data center by 2025), the picture is still unclear. Sangam outlined the evolving dynamics between mobile network operators and hyperscalers.
“Edge computing is where these two ecosystems meet, collide or compete,” he said. “They have to work together.” The major hyperscaler is currently notable for the private sector, although it’s worth noting that he appears to be seizing the opportunity at the confluence of networking and edge computing. looks like the operator is in control.
Another question raised by Sangam worth considering is the need for multi-operator, multi-cloud deployments. For example, having one cloud provider and her one operator working together can be problematic for companies operating in multiple countries. Complications can also arise when one operator works with one of her cloud providers and that cloud provider is not the best partner for the end user. “It has to be multi-cloud and multi-operator. It’s complicated too,” he admitted.
In short, Sangam characterized mobile edge computing as bringing a “middle tier” that avoids the cost and time of moving data back to the cloud for processing. Leveraging that middle-tier intelligence enables faster action based on data intelligence. “Whatever you call it, it’s very clear from industry trends that value and a lot of intelligence is moving to the edge of the network…immediate processing and intelligence happens in the edge cloud.”