Modern digital infrastructure

Infrastructure is the underlying physical and virtual equipment that lets data flow: cables, routers, servers, towers and the services run on them. Think of roads and bridges: the traffic on them (your data) can be anything, but how well it moves depends on the roads. This section covers five pieces of modern infrastructure. For each one you should be able to say what it is for, what it is good at, and where it falls down.

The internet is a global network of networks. Millions of private, public, business and government networks are joined by routers and speak the same protocols, so any device on one can reach any other. The web is just one service running on top of it, alongside email, video calls and online games.

• Purpose: reliable, fault-resistant communication and data exchange across the world.
• Benefits: instant global communication, access to enormous amounts of information, cheap commerce and collaboration, shared teaching and learning.
• Limitations: speed varies with distance, congestion and local infrastructure; it must keep growing without slowing down; and its openness is a security risk to people, companies and nations.

Cloud computing (A1.1.9 covers its service models) means renting remote servers, storage and software over the internet instead of owning them. A school can store files and run its email in the cloud without running a server room.

• Benefits: scale up or down within minutes, pay for what you use, no hardware to maintain, access from anywhere, built-in backup and disaster recovery.
• Limitations: useless without a good connection; data sits with a third party (security, privacy and legal rules); heavy data transfer can be expensive; and the physical distance to the data centre adds latency for real-time work.

A distributed system is a group of independent computers (nodes) that pass messages to each other and look to the user like one system. Search engines, online games and big-data tools like Hadoop work this way.

• Purpose and benefits: share the work, so jobs finish faster; add nodes to grow; keep going when a node fails, because no single machine is essential.
• Limitations: keeping every copy of the data consistent is hard; coordinating nodes over an unreliable network is complex; there are more machines to attack; and messages between nodes add delay.

Edge computing moves the processing to where the data is produced (the “edge” of the network) instead of sending everything to a distant cloud. A smart traffic light that adjusts to the cars it sees, or a camera that reads number plates and sends only the plate text, are edge systems. Streaming companies also place cache servers in cities so popular films come from nearby.

• Benefits: much lower latency (vital for real-time decisions), less bandwidth used, and sensitive data can stay on site.
• Limitations: many scattered devices to secure, update and manage; higher set-up cost; devices of uneven power; and keeping data consistent across them.

A mobile network is a cellular network: the land is divided into cells, each served by a tower, and your connection is handed from cell to cell as you move. It gives voice, messaging and data without a cable, and is the only connection in many parts of the world. From 2G to 5G, each generation has raised speeds and cut latency.

• Benefits: connectivity almost anywhere, mobility, and a whole ecosystem of apps (payments, maps, telemedicine, remote work).
• Limitations: dead zones; quality that changes with distance, obstacles and crowds; radio signals that can be intercepted; and fake towers or hotspots that lure devices in.