The key goal of edge computing, an emerging sector, is to provide faster service of online content to a user population with reliability and enhanced connectivity based upon moving the servers in closer proximity to the user and the data. To those who work in the edge computing sector, it would be helpful to review similar issues that have arisen in the distributed energy generation (DG) industry, which locates small generators near or within the facilities of users. This article focuses on conceptual parallels between the DG and edge data center sectors. Where there are parallels, there are also lessons to be learned.
The paramount driving concept in each sector is that there are benefits associated with locating the service provider close to the user and the source of energy (with DG) or the data (with edge computing).
In DG, since the generation facility is located at or proximate to the facility that uses the power, transmission losses are reduced, requiring less transmission infrastructure. DG can also be interconnected with the electric utility grid so that when the DG is not operating, power can be supplied from the grid. In addition, when the grid is not operating, certain types of DG can provide power.
Similarly, in edge computing, a multi-access edge computing (MEC) service and application node is located at or proximate to both the service user and data source, and it is also interconnected with other internet and cloud resources. Edge computing centers may be small and managed remotely, but they are rarely standalone. Typically, they are managed from a centralized location that functions as a hub. To extend the parallel between DG and edge computing further, there are as many instances of utilities building DG in addition to their central power plants as there are data center and colocation providers enlarging their edge position by positioning new or acquired data centers to expand their edge footprints.
The similar benefits of these two technologies is that each:
- Shortens transmission distances, which reduces the costs for acquiring or building energy/data transmission infrastructure
- Cuts down on network congestion
- Reduces inefficiency
- Aids reliability by reducing reliance on central facilities
- Improves security
- Permits more flexibility in structuring networks
- DG reduces power losses and, as to edge computing, reduces duration of transmission time; and, in each case, reduces costs
DG requires a two-party contract between a third-party DG provider (DG provider) in the form of, e.g., a power purchase agreement. Similarly, the relationship between the edge computer data provider (edge provider) and an edge user (edge user) may also involve a provider/a two-party edge user contract — unless the edge user is a mobile edge user, in which case, there will likely be a clickwrap form of agreement.
DG provides definite benefits, but also involves certain legal and financial risks, and certain similar benefits and similar risks also apply to edge computing.
Real Property, Lease, and License Considerations
Acquiring real property rights. “The edge is not a technology land grab,” remarked Cole Crawford, CEO of µDC producer Vapor IO. “It is a physical, real estate land grab.” While “grab” may be an overstatement, the gist of Mr. Crawford’s comment is correct — the edge requires real estate acquisitions, installation of equipment, and facilities and related rights. Years of experience with the utility and DG industry has taught us that acquiring real property rights to construct and install facilities can be time-consuming and expensive to locate, negotiate, purchase, or lease. In addition, edge providers do not have the right of condemnation enjoyed by utilities which, in certain cases, might enhance their roll-out of edge facilities.
Edge providers face issues similar to DG providers. DG generators and related facilities must be installed, which may require transfer of real property, leasing, or licensing. Financiers typically require “site control” sufficient so that facilities used for collateral can be removed. Yet mortgage holders for facilities can sometimes refuse to permit DG financiers to have “prior” lien rights for the installed equipment — even when the owner of the facility wishes to have DG installed. Other issues can include sale of the facility in which the DG is located or destruction to the facility. Edge providers can expect to run into the same problems since the installation of server equipment in numerous sites will require transactions with multiple entities in many locations.
Utility poles and towers. Like the CATV industry, edge providers may seek to install certain equipment on utility poles and towers or the related right-of-ways. Utility poles and towers can be crowded and hotly contested “real estate.” Witness the experience of the CATV industry seeking space on electric and utility poles to extend their facilities. That industry’s need for locating equipment on utility poles led to years of proceedings before the public service/utility commissions in various states. There are also space, cost, anchoring support, and safety distancing considerations to mounting equipment on utility poles. Those seeking to locate facilities on utility right of ways also face obstacles.
Performance liability. Performance liability issues relating to DG provider obligations offer guidance for edge providers. Small and dispersed DG installations are typically remotely monitored but do not have continual on-site supervision and maintenance. This same situation also exists for dispersed edge installations that are scattered and remote. Monitoring, maintaining, and performing emergency services for a DG power plant or servers in a large data center or other continually staffed facility is much more readily accomplished than performing care in scattered sites. DG providers promise certain performance levels — and any failure to meet those levels may trigger contractual liability.
Similarly, the promise of edge computing is faster, more accessible, and more reliable access. Edge users will likely pay more for this service and will receive specific promises of performance in return for their payments. If the promised level of service is not provided, edge providers could face liability. For edge providers serving financial industry users, that liability could be significant. For that reason, the covenants addressing performance and related remedies should be carefully drafted. Further, for edge providers providing services to entities involved in the growing blockchain services market, there may well be additional risks for providing less than the promised level of service. (Note: The possible effects on edge computing and the pricing of its services from the scheduled demise of net neutrality are beyond the scope of this article.)
Counter-party risks. Each party must consider the risks of contracting with the other party in terms of reliability and capability to meet possible liabilities. In entering into any agreement, it is important to consider not only the remedies if the other party fails to fulfill its obligations under the agreement but also the ability of the other party to be able to perform. In this article, we are not considering circumstances in which the counter-party is excused from performance by events beyond its control (weather events, international crises, etc.). An edge provider may go over the “edge” and promise more than it can reasonably expect to deliver in terms of reliability or speed in certain cases. The possibility of the edge provider failing to perform as promised or failing to perform at all must be considered by the edge user — and its lawyer — before the contract is signed. Advance due diligence, including financial due diligence, is a must.
Warranties. New and often technologically advanced equipment is typically installed at data centers, whether DG or edge data centers. But what warranties will be given for performance? We have found that, in DG installations, the warranties may be limited to repair and replacement, which may be insufficient to permit the DG provider to, in turn, fulfill any failed obligations to the DG users. This may become an issue for edge users. For example, unscheduled downtime may require the DG user or edge user to seek service from another provider at greater cost. The providers must take care that that they do not become caught in the squeeze of having made promises to their customers that are not covered by the equipment warranties the providers have received from equipment vendors.
Allocation of rights and risks. The allocation rights and risks in DG agreements can be readily determined as can the rights and risks in bilateral edge agreements between two entities. As to other forms of agreements which may be applicable in DG and edge computing, we note that utility tariffs and clickwrap agreements, unfortunately, offer scant opportunity for reasonable review. A utility tariff is often hundreds of pages with extensive internal cross-references and can contain difficult-to-understand language which is legally imposed upon all users in the service territory. Clickwrap agreements, which most users typically accept without examination (despite the admonition to “carefully review”), can also be surprisingly long and one-sided. The opportunity for an individual user to effectively challenge a tariff or a clickwrap agreement is next to nonexistent, although courts are beginning to impose limitations.
To conclude, edge users and providers can anticipate issues and problems by reviewing the way the DG industry has addressed property rights and negotiated contractual language. As the edge industry grows, and issues and problems multiply, edge users and providers can look for lessons to the DG industry.
- “Where’s the ‘edge’ in edge computing? Why it matters, and how we use it.” B Scott Fulton, ZD Net. April 17, 2008.