Understanding Infrastructure Edge Computing. Alex Marcham

topology, as from its routing table.

      2 Before the traffic can be sent, the layer 2 address of the next hop destination for the traffic must be determined. Remember that in the OSI model, as we send traffic we must traverse down the stack towards layer 1. We cannot simply skip layer 2 just because we are using layer 3 addressing as well in the network. Our IP packet at layer 3 will first be encapsulated in an Ethernet frame at layer 2, which needs a pair of source and destination MAC addresses. The source MAC address is that of the interface about to send the traffic on to its destination; but the destination MAC address may not be known yet.

      3 The endpoint checks its switching table, also referred to as its Address Resolution Protocol (ARP) table in the case of IPv4. This table contains a list of MAC addresses that are matched with IP addresses. If an entry exists that matches the destination IP address of the traffic to a MAC address, that MAC address is then used as the destination MAC address of the Ethernet frame being created. If it is not known, the ARP protocol (or an equivalent process based on the specific layer 3 protocol in use) is invoked to discover that destination MAC address.

      4 With the IP packet encapsulated in an Ethernet frame with both source and destination MAC addresses, the Ethernet frame is then ready to be transmitted to its destination. Where layer 2 switching occurs, only the layer 2 information contained in the frame is required; and when a device uses layer 3 information to perform routing, it decapsulates the Ethernet frame and then acts upon the layer 3 information of the IP packet within. When traffic must then be sent towards its next hop destination, this encapsulation process is repeated.

      In this way, layer 2 and layer 3 technologies function together closely to transport traffic from its source to its destination across a network that may span from one end of the same building, or it may span the globe in the case of an application delivered across the internet. Regardless, the same basic processes are repeated to move traffic over the network irrespective of any physical distance.

The examples in this section are a key foundation for topics in later chapters, which will use network virtualisation and inter‐layer interoperation to provide the flexibility and performance required to support next‐generation infrastructure and applications. Although many of the characteristics of and use cases enabled by infrastructure edge computing are new, these same basic processes apply from a network infrastructure and operation perspective just as they do to the current networks of today.

3.7 LAN, MAN, and WAN

      Modern networks exist at drastically varying sizes, and it is useful to categorise them into three main classes according to their scale. The three most commonly used terms to describe the scope or scale of a particular network are the local area network (LAN), metropolitan area network (MAN), and wide area network (WAN), in order of their increasing geographical size. Although there are no real hard and fast standards that dictate the size a specific network must be to qualify as a specific scale denoted by one of these terms, it is usually not difficult to come to an agreement on terms to use.

      In many cases, there is an hierarchical relationship between these three grades of network scale. One LAN may be combined with many others in a metropolitan area and can be interconnected to form a MAN; and one or more MANs across multiple metropolitan areas may be interconnected to form a WAN. On the other hand, networks at any of these three scales could be created as single networks. This choice is driven by a combination of business and technical factors depending on the individual parties and technologies involved in a specific area and is not prescribed by the terms themselves.

To show each of these network types visually, consider the diagram in Figure 3.3, which shows two cities:

Figure 3.3 LAN, MAN, and WAN networks.

The diagram shows multiple LANs within a city, a single MAN covering all of that city, and a WAN that is connecting the two cities to each other despite them being 100 miles apart. Of course, a city may have hundreds or thousands of LANs, and there may be multiple MANs and also multiple WANS in that area or between areas; but this example serves to show the difference in scale between typical networks in each of these categories and how one may appear to nest within another from above.

      There is also not necessarily any direct hierarchical relationship between these network categories at all. For example, depending on the network topology in a particular area, a LAN may just connect directly to a WAN. In another, it may need to connect to a MAN, which itself connects to a WAN. The specifics in this regard are location and implementation choices made by those network operators.

      As well as a different physical scale, the purposes of each of these types of networks are different from one another. A LAN is typically used to connect endpoints within a single building or campus together or to other network resources, whereas a MAN is often used to connect multiple LANs to each other across an area such as a city. A WAN, then, is typically used to connect one network or endpoint to a network resource that is a significant distance away, hence its “wide area” naming.

3.8 Interconnection and Exchange

      As mentioned previously, the internet is far more than the combination of physical infrastructure and logical protocols; it is a network of networks that are joined as much by agreements between their operators towards a mutual benefit as they are by anything else. This can be best seen in the way in which the networks of different operators interconnect and exchange data. This exchange of data between networks is essential; without it, endpoints of one network would not be capable of accessing resources or endpoints on another network, making the internet we know today useless.

      The large‐scale exchange of data between networks typically occurs inside an internet exchange (IX). Often, IXs are located within large data centres which terminate a high number of WANs inside the facility. The higher the number of networks aggregated into one location, the higher the theoretical value of that IX as a place for networks to exchange data. Chapter 7 focuses on this interconnection.

The physical locations of IXs are of considerable interest when thinking about infrastructure edge computing. Our previous example describes only the location of the IX itself, and not the locations of any of the endpoints which are ultimately sending and receiving the traffic. Consider the scenario, however, where these two endpoints are located close together but neither is near to an IX. In this example, each endpoint is connected to a different access network, and so for the data that they are sending each other to reach its destination, it must go all the way to the IX, be exchanged between their networks in that remote location, and then it will be sent all the way back to those endpoints.

      Where the endpoints are using applications that rely on either very low latencies between endpoints or are sending a large amount of data, this is not an ideal network topology. For the former, latency is added between the source and destination of the traffic due to its need to traverse the IX before it can be delivered to its target endpoint, which would be minimised if a point of data exchange were available closer to those endpoints. For the latter, network resources are used between each of the endpoints and the IX to transfer the traffic, which some entity (even if it is not the users of each of the endpoints themselves directly) must pay for, with the possibility to introduce congestion as well.

      These factors have led to the need for more distributed points of data exchange across the internet. Infrastructure edge computing addresses this challenge by the implementation of an edge exchange (EX) within infrastructure edge data centres. Conceptually, this idea is simple; a smaller‐scale version of the functions performed by an IX can be implemented inside an infrastructure edge data centre to allow networks to exchange data at locations closer to the endpoints, which are generating the traffic in order to decrease latency and minimise