Let's start with the fact that there is not a big difference between an edge server and a standard bare metal server. However, there is a huge difference between the application performance of a standard dedicated server and an edge server, even if both machines have the same amount of computing resources.
Two things make the difference - the data center location and the network. For a dedicated server in New York to be considered an Edge one, it has to be physically located in a New York-based data center or in one that is very close to the city. The second important thing is the network latency to the local users. It must be less than 5 milliseconds, the lower the better. The best network latency for an Edge server for local application delivery would be 1 ms on average.
An Edge server is a great solution for hosting various private workloads. Organizations have traditionally done this in their offices, something that has begun changing with cloud computing. The major clouds like AWS, Azure, Google (News - Alert), and other big cloud hosting providers managed to successfully sell the idea that it is better for organizations to have their technology environments hosted in the cloud than keep them in an on-premise environment. This is partially true as if set properly, the cloud-hosted environments are more resilient and offer business continuity, something that is much harder to achieve in a local office environment due to various factors - connectivity, lack of power redundancy, hardware depreciation, and many other factors.
The big cloud providers, however, have at least three major disadvantages when it comes to hosting private infrastructure. In many scenarios, they feature a longer network latency as their data centers are often outside the main metropolitan markets, and they charge their clients huge fees for data transfer. Add to that the fact that your data hosted on any of the major clouds is technically not private, so it is not your data. If an organization hosts its application on its own server colocated in any data center, then the data is 100% private. It is a private property. When hosted on the cloud, the data is hosted on an infrastructure that belongs to the cloud service provider. Therefore, various terms of service apply.
With the dedicated servers and Edge Servers in particular, it is a matter of an agreement. As these are physical appliances used on leased contracts, they are owned by the hosting providers. However, the organizations that use them can always request a special clause in the contract that specifies that whatever is hosted on the bare metal servers, is owned by the organization is considered private property, and may be touched only if there is a court order. So using a bare metal server is a much better option than using an infrastructure-as-a-service (IaaS) delivered by any big cloud provider.
The IT department of any small or middle-sized organization should be able to plan the technology workloads and set up a dedicated Cloud infrastructure on any bare-metal servers, rather than on any major cloud. The bare metal nodes can be customized with any virtualization technology and when used on a month-to-month contract, the organization could scale the resources up and down, based on demand. Even, if the resource planning cannot be that precise and even if the IT departments need to overprovision, it is still worth it for various reasons.
Financially-wise it is much better to create a dedicated cloud infrastructure on bare-metal servers, instead of using Public or Private infrastructure with the big cloud providers. The reason is that the bare-metal servers can be provisioned with high-bandwidth rate ports. Every node could be a 10Gbps dedicated server, for example. If the organization needs to create a cluster of 3, 4, or five bare-metal servers, which would have plenty of bandwidth and would never need to pay for overage data transfer, for DNS zones, and will avoid the perplexed billing introduced by the major clouds.
In terms of overall service quality, the bare-metal servers feature infrastructure technical support, something that organizations are forced to pay separately if they lease IT infrastructure from AWS, Azure, or any other major cloud.
The paradigm of Edge computing extends The Cloud to various edge locations and gives a chance for smaller data centers and edge computing service providers to cope with the major clouds, which own and operate large data center footprints. The mobility of the infrastructure operations and the ability to deploy computing capacity fast where there is a demand for instant application delivery, are the two most important factors that drive the edge infrastructure service market forward.
In the Edge Hosting market, the key is not the ability of the Cloud provider to instantly scale up the IT infrastructure. Scalability is still very important on the edge, but it is not a defining feature. The main purpose of the edge servers is to store data and process data on-site, where it is collected or as close as possible to the point where certain events occur. For example, various application services related to road traffic are collected on-site, stored, processed, and analyzed in real-time in the edge data centers and networks that offer 1ms - 3ms latency to the data usage device or the users.
As a technology trend edge computing and edge hosting are also driven by decentralized computing infrastructures and blockchain-based operations. We are still at the very beginning of the development of the decentralized digital economy, driven by individuals, not by corporations, in which the business users purchase various technology services with digital currencies, for example, to buy dedicated server with Bitcoin from platforms and decentralized marketplaces, rather than from any of the big cloud infrastructure providers.
This is where the small and medium data center service providers have a strategic advantage. They can plan and deploy small data centers or even mobile containers for hosting IT equipment. The new SMB infrastructure providers that provide services on the edge are not focused on the global market like the traditional web hosting provider, they do business locally. Their business success depends on their ability to communicate and persuade the local businesses and organizations that store and process data on the edge and deliver application services with the shortest possible delay while improving the quality of the technology service, but it also saves financial significant resources.
As with many other new computing trends that bring decentralization to the IT service market, Edge computing is exciting. Its infrastructure service layer, however, is just the ground floor of a huge virtual marketplace for digital services. There are many other service layers, such as networking, software development automation, and many others that create huge opportunities for hundreds of millions of people.
