Edge computing challenges
While edge computing has the potential to offer compelling benefits across a variety of use cases, the technology is far from foolproof . In addition to traditional network constraints, there are several key considerations that may impact the adoption of edge computing：
- Limited ability . Part of the appeal that cloud computing brings to edge (or fog) computing is the diversity and scale of resources and services. Deployment of infrastructure at the edge can be effective, but the scope and purpose of the edge deployment must be clearly defined - even extensive edge computing deployments use limited resources and few services to achieve a specific scale at a predetermined scale. Purpose
- Wired . Edge computing overcomes typical network limitations, but even the most forgiving edge deployments require some minimum level of wiring. It is important to design a maladaptive or disconnected edge deployment and consider what happens to the edge when connectivity is lost. Autonomy, artificial intelligence, and elegant failure planning after wiring issues are important for successful edge computing.
- Safety . IoT devices are notoriously insecure, so it is important to design an edge computing deployment that will emphasize proper device management, such as policy-driven configuration enforcement, and security of computing and storage resources - including software patches and updates. factors – and pay particular attention to encryption at rest and in-flight data. IoT services from major cloud providers include secure communications, but this is not automatic when building an edge site from the ground up.
- Data life cycle . The long-term problem with today’s data glut is that much of it is unnecessary. Consider a medical monitoring device - it's only the problem data that matters, and there's little point in storing a few days of normal patient data. Most of the data involved in real-time analysis is short-term data and is not stored long-term. Once analyzed, businesses must decide which data to keep and which to discard. Retained data must be protected in accordance with business and regulatory policies.
Edge computing implementation
Edge computing is a straightforward idea that may look easy on paper, but developing a cohesive strategy and implementing a sensible deployment at the edge can be a challenging endeavor.
The first key element of any successful technology deployment is establishing a meaningful business and technology advantage strategy . Such a strategy is not about cherry-picking suppliers or gear. In contrast, edge strategies take into account the needs of edge computing. Understanding the "why" requires a clear understanding of the technical and business issues that organizations are grappling with, such as overcoming network constraints and adhering to data sovereignty.
The first key element of any successful technology deployment is establishing a meaningful business and technology advantage strategy . Such a strategy is not about cherry-picking suppliers or gear. In contrast, edge strategies take into account the needs of edge computing. Understanding the "why" requires a clear understanding of the technical and business issues that organizations are grappling with, such as overcoming network constraints and adhering to data sovereignty.
This strategy might start with a discussion of what edge means, where it exists in the enterprise, and how it should benefit the organization. Edge strategies should also be consistent with existing business plans and technology roadmaps. For example, if an enterprise is looking to reduce its centralized data center footprint, edge and other distributed computing technologies may align well.
As the project approaches implementation, it is important to carefully evaluate hardware and software options. There are many vendors in the edge computing space , including Adlink Technology, Cisco, Amazon, Dell EMC and HPE. Each product must be evaluated for cost, performance, functionality, interoperability and support. From a software perspective, tools should provide full visibility and control of the remote edge environment.
Actual deployments of edge computing initiatives can vary widely in scope and scale, from some local computing on top of a utility or device in a ruggedized enclosure to a large number of sensors providing high-bandwidth, low-latency network connectivity to the public cloud. . No two edge deployments are the same. It is these changes that make edge strategy and planning so important to the success of edge projects.
Edge deployments require comprehensive monitoring. Keep in mind that getting IT staff into physical edge sites can be difficult or even impossible, so edge deployments should be designed to provide resiliency, fault tolerance, and self-healing capabilities. Monitoring tools must provide a clear overview of remote deployments, enable easy provisioning and provisioning, provide comprehensive alerts and reporting, and maintain the security of the installation and its data. Edge monitoring typically involves a range of metrics and key performance indicators , such as site availability or uptime, network performance, storage capacity and utilization, and computing resources.
No edge implementation would be complete without careful consideration of edge maintenance:
As the project approaches implementation, it is important to carefully evaluate hardware and software options. There are many vendors in the edge computing space , including Adlink Technology, Cisco, Amazon, Dell EMC and HPE. Each product must be evaluated for cost, performance, functionality, interoperability and support. From a software perspective, tools should provide full visibility and control of the remote edge environment.
Actual deployments of edge computing initiatives can vary widely in scope and scale, from some local computing on top of a utility or device in a ruggedized enclosure to a large number of sensors providing high-bandwidth, low-latency network connectivity to the public cloud. . No two edge deployments are the same. It is these changes that make edge strategy and planning so important to the success of edge projects.
Edge deployments require comprehensive monitoring. Keep in mind that getting IT staff into physical edge sites can be difficult or even impossible, so edge deployments should be designed to provide resiliency, fault tolerance, and self-healing capabilities. Monitoring tools must provide a clear overview of remote deployments, enable easy provisioning and provisioning, provide comprehensive alerts and reporting, and maintain the security of the installation and its data. Edge monitoring typically involves a range of metrics and key performance indicators , such as site availability or uptime, network performance, storage capacity and utilization, and computing resources.
No edge implementation would be complete without careful consideration of edge maintenance:
- Safety . Physical and logical security precautions are important and should include tools that emphasize vulnerability management and intrusion detection and prevention. Security must extend to sensors and IoT devices because each device is an element of the network that can be accessed online or hacked—presenting a bewildering number of possible attack surfaces.
- Wired . Wiring is another issue, and provision must be made for access control and reporting even if wiring to the actual data is not available. Some edge deployments use secondary cabling for backup cabling and control.
- Manage . Edge deployments are remote and often unavailable, making remote configuration and management very important. IT managers must be able to see what is happening at the edge and be able to control deployments when necessary.
- Physical maintenance . Physical maintenance requirements cannot be ignored. IoT devices often have a limited lifespan, requiring regular battery and device replacement. Gears fail and eventually require maintenance and replacement. Maintenance must include actual on-site logistics.
The possibilities of edge computing, IoT and 5G
Edge computing continues to evolve, using new technologies and practices to improve its capabilities and effectiveness. Perhaps the most noteworthy trend is edge availability, with edge services expected to be available globally by 2028. Where edge computing is often situation-specific today, the technology is expected to become more ubiquitous and change the way the internet is used, bringing more abstractions and potential use cases to edge technology.
This can be seen in the proliferation of computing, storage, and networking device products designed specifically for edge computing. More multi-vendor partnerships will enable products to achieve greater interoperability and flexibility at the edge. One example includes the partnership between AWS and Verizon to provide better connectivity edges.
Wireless communications technologies, such as 5G and Wi-Fi 6, will also impact edge deployment and utilization in the coming years, enabling unexplored virtualization and automation capabilities such as better vehicle autonomy and workload migration to the edge, while Make wireless networks more flexible and cost-effective.
This can be seen in the proliferation of computing, storage, and networking device products designed specifically for edge computing. More multi-vendor partnerships will enable products to achieve greater interoperability and flexibility at the edge. One example includes the partnership between AWS and Verizon to provide better connectivity edges.
Wireless communications technologies, such as 5G and Wi-Fi 6, will also impact edge deployment and utilization in the coming years, enabling unexplored virtualization and automation capabilities such as better vehicle autonomy and workload migration to the edge, while Make wireless networks more flexible and cost-effective.
With the rise of the Internet of Things and the sudden glut of data generated by such devices, edge computing is attracting attention. But with IoT technology still in its relative infancy, the evolution of IoT devices will also have an impact on the future development of edge computing. An example of such a future alternative is the development of Micro Modular Data Centers (MMDCs). An MMDC is basically a data center in a box, putting a complete data center in a small mobile system that can be deployed closer to the data-such as across a city or region-to bring computing closer data without placing edges on the data itself.