Convenience computing offers the robustness of central data centers at the edge. Let’s take a look at this option.
Too often Internet of Things A device that runs a manufacturing production line, robotics, sensors, and other IoT devices operate strictly on distributed networks. Distributed networks may need to connect to a central data center to perform heavier processing. However, proximity computing is a distributed data center and IT infrastructure that can optimally support the IoT implemented in different physical locations.
see: Increased adoption of advanced computing by 2026. Organizations are cautious about adding 5G to the mix (TechRepublic Premium)
The concept behind Proximity Computing is that resources previously hosted in corporate data centers can actually be moved to edge micro data centers. This enables more robust computing at the edge that can perform many functions previously reserved for central data centers, and supports IoT better than today’s widely deployed standard networks. to augment.
Whether you decide to move your physical micro-datacenter to the edge, use a cloud-based mini-datacenter, or use a combination of the two, you’ll end up with an on-premises tier. Scalable processing and storage suddenly becomes available. IoT applications in manufacturing plants and remote offices. This eliminates the pressure of doing all the work from a central data center and reduces the costs associated with sending data over communication lines.
“Proximity computing is primarily powered by micro-modular data centers and software-defined fiber optic networks,” said Cole Crawford, CEO of VaporIO, an edge colocation and interconnect service provider. “Optical networks create a flat Layer 2 network that connects the company’s facilities to an on-site data center and makes the data center equipment behave as if it were in the field,” because the on-site data center is less than 10 miles away. [IoT] At the target site, the network can reach a delay of 75 µs. This is a level of performance that makes offsite equipment behave as if it is onsite. “
see: The future of agriculture: AI, IoT, drones, etc. (Free PDF) (TechRepublic)
For IoT initiatives like Manufacturing 4.0 that rely on sensors, robotics, and equipment requiring interaction and interaction, mini-data centers that deliver unmatched scalability and agility at the edge can revolutionize. ..
The question is whether the CIO sees local IT as transformative.
“Real change is cultural,” Crawford said. “Businesses must first adopt a mindset that recognizes the benefits of moving or expanding their on-premises infrastructure to a closer location. Closer to places is the new best for deploying private cloud workloads. Should be considered a practice. “
Of course, this requires changes to the IT infrastructure, such as redeploying work processes and storage. As IoT is established at the edge, CIOs and system architects recognize this trend, but so far they have approached it with standard network deployments.
see: Three Benefits of Having Cloud Tools Available in Your On-Premises Data Center (TechRepublic)
Strategically, considering hybrid, on-premises or near-cloud data centers is not a big step. However, there are still issues such as how to redeploy under budget constraints and knowing which existing resources should continue to operate. CIOs and infrastructure architects also need to find the time to rebuild their IT infrastructure for nearby IT.
Crawford claims that companies that embrace convenience computing will reduce the total cost of ownership of their compute and storage infrastructure by 30-50%, thereby eliminating most or all of the capital costs that would normally be spent on computing and storage. data center itself. I said I can. These benefits can also be exacerbated by turning capital expenditures into operating costs through a new, scalable service model.
If CIOs can demonstrate these benefits in a cost model prepared for their IT budget, local IT could certainly be a new strategy for implementation at the edge.
Don’t overlook the resiliency of local computing. “While the performance of on-premises computing is comparable to that of on-premise computing, it also has the ability to dramatically improve resiliency,” Crawford said. “Businesses can spread their workloads across multiple nearby facilities, eliminating single points of failure and enabling software-based, high-availability resiliency technology. ”
There is an average production line failure of $ 17,000 per incident, manufacturer with an average of 800 hours of downtime per year System resiliency, agility, scalability and robust security are important considerations on the list of CIOs and are the reasons for serious consideration of near-site computing at the edge.