• State of the Edge Report 2023 (The Linux Foundation)

  (https://stateoftheedge.com/reports/state-of-the-edge-report-2023/)

• From DevOps to EdgeOps: A Vision for Edge Computing (Eclipse Foundation) (https://outreach.eclipse.foundation/edge-computing-edgeops-white-paper)
• Sharpening the Edge – Part 1 (LF Edge) (https://lfedge.org/wp-content/uploads/sites/24/2023/12/LFEdge_Akraino_Whitepaper2_v1_PrePrint.pdf)
• Sharpening the Edge – Part 2 (LF Edge) (https://lfedge.org/wp-content/uploads/sites/24/2023/12/LFEdgeTaxonomyWhitepaper_062222.pdf)
• Software-as-a-Service (SaaS) overview (https://www.salesforce.com/saas/)
• Defining software deployment as days (https://dzone.com/articles/defining-day-2-operations)

In this section, you will learn to articulate fundamental differences between the edge and the cloud. This impacts the available infrastructure, platforms, services, and application deployments. Additionally, you will be able to explain concisely what the edge is and what the field of edge computing does to both line-of-business (LOB) executives and other non-IT professionals. This includes being able to explain the value it can provide, as well as why this field is emerging at this point in time.

What is the edge?

The edge in edge computing is commonly used to describe the location where computing takes place. The name itself is meant to evoke a spot in a corner or by the wayside, and not in a central area, and actually refers to the very end of a communications network (the edge of the internet; see Figure 1.1). Thus, edge computing happens outside a cloud computing facility, and many times outside the four walls of a traditional data center (DC...

In this section, we cover the names and characteristics of various edge categories (or edges) that are commonly used, including which terms have been deprecated or have fallen out of the vernacular. By the end, you should be able to list the edges and describe the benefits and drawbacks of each, as shown in exhaustive detail next (Figure 1.3):

Figure 1.3 – Detailed benefits and drawbacks of each edge

Image Source: Linux Foundation Whitepaper

The user edge – field-deployed compute

A fundamental confusion that commonly comes up in conversations revolves around where people think the edge is. One idea in people’s minds is that “the edge” may be in houses, commercial offices, factories, vehicles, and utility shacks on the side of the road or at the base of a cell tower. These types of locations are typically thought of as the far edge because they are farthest away from a DC or cloud on...

In this section, we review different scenarios where more than one application or service can be deployed and running concurrently. We will give an example of each and discuss the purpose of that approach. By the end, you should recognize when each tactic is required.

Edge computing doesn’t require dedicated resources

In Sharpening the Edge, the author refers to edge computing as primarily using your own systems, while cloud computing involves sharing systems and infrastructure with others. This simplification is largely correct at the macro scale, although edge computing can also involve sharing: by running multiple applications on a single device at the same time for multiple users, or sequentially at regularly scheduled intervals (day versus night, weekdays versus weekends, business open versus business closed). In the case of CSP-hosted edge infrastructure, it could even include hosting applications from multiple...

Another common way of describing an architecture is based on its foundation and assumptions, and in what direction and manner it grows as its scope and responsibilities increase. If it starts out based in the cloud, using cloud-native best practices, and then later adds on capabilities that allow it to run in some fashion on the edge, that approach is described as cloud-out. On the other hand, if it starts out on the edge using edge-native development best practices, and then bursts to (hybrid) cloud infrastructure, that would be termed edge-in. Let’s look at each in turn, go over an example, and discuss their relative strengths and weaknesses.

Looking deeper at cloud-out architectures

Cloud-out architectures begin with the cloud; that is, with global compute infrastructure using cloud-native development best practices. While Chapter 11 will discuss these practices in depth, let’s list an abbreviated version to frame the discussion:

...

In this section, we introduce you to the concept of archetypes. Along the way, we cover the days involved in the software life cycle and discuss deployment methods. By the end, you will be ready to learn about and use archetype patterns.

What is an archetype?

An archetype is the original model or form of something that embodies all primary qualities of that item, whether that item is an abstract concept or a physical object. In the case of architectural patterns, our assertion is that most application architectures can be derived from an original pattern archetype (source) or a slight variation based on local or business considerations. Therefore, in this book, we attempt to tease out archetypes and discuss them at length based on our belief that identifying and mastering them will give you the foundational skills needed to tackle most, if not all, edge application scenarios. By following these patterns judiciously, you will be able to create solutions...

In this chapter, you learned about what capabilities the edge and the cloud have in common, and what distinguishes one from the other when it comes to the tiers of compute: infrastructure, platforms, services, and applications.

We covered the names and characteristics of edge categories and described the benefits and drawbacks of each. We also discussed various ways that edge nodes can be shared between tasks. We described how an architecture can be scaled based on its scope and responsibilities with cloud-out and edge-in paradigms. Also, we introduced you to the concept of archetypes and the “days” involved in the software development life cycle.

Now, you should be ready to learn about the basic components and building blocks that go into archetype patterns. Just as importantly, we’ll also discuss how to approach solutions from the right perspective in order to build a future-proof application, follow best practices, and use long-term thinking.