Computer Science & Algorithms
Section 1
Core Idea
Packet switching sends data in small, discrete packets that can be routed independently across a shared network. Unlike circuit switching (dedicated path for the whole session), packets from many flows share links; each packet carries addressing info and is forwarded hop-by-hop. The network can use capacity flexibly—no single call monopolises a path—so aggregate throughput scales with statistical multiplexing. Tradeoff: variable latency and need for buffering and congestion control when many flows compete. The basis of the internet and most digital communication; the same idea applies to task queues, batch jobs, and any system that multiplexes work over shared resources.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Section 2
How to See It
Building & Scaling
You're seeing Packet Switching when work or data is broken into chunks that share capacity and are routed or scheduled independently—message queues, micro-batches, API requests, or event streams. Scaling comes from multiplexing, not dedicated pipes.
Section 3
How to Use It
When demand exceeds a single path's capacity, packetise: break work into units, add routing or routing keys, and let a shared fabric deliver them. Design for variable latency (buffers, timeouts) and congestion (backpressure, load shedding). In product and ops: think in "packets" of work that can be distributed and parallelised instead of one long-lived pipeline.
Decision filter
"Do we have shared capacity and bursty or diverse demand? If yes, packetise work so it can be multiplexed and routed; accept variable latency and design congestion control."
As a founder
Product and eng decisions often mirror packet switching: break big initiatives into shippable units, route them to the right team or queue, and let the system multiplex. Avoid dedicating the whole "circuit" to one initiative; packetise so capacity is shared and throughput scales.
Section 5
Founders & Leaders
Reed HastingsCo-founder & CEO, Netflix
Hastings built Netflix on streaming over the internet—packet-switched networks. Content is delivered in chunks over shared infrastructure; adaptive bitrate and buffering handle variable latency. Founders can take the lesson: design for shared, multiplexed capacity and variable conditions; packetise work and data so the system scales and degrades gracefully.
Section 7
Connected Models
Reinforces
Modularity
Packet switching is modularity in the network: each packet is a unit; routing and forwarding are standardised. Building systems from small, routable units (packets, tasks, events) is the same principle applied to architecture.
Tension
Theory of Constraints
TOC finds the bottleneck and improves it. Packet switching spreads load across the network; the bottleneck may shift (e.g. a congested link). The tension: multiplexing increases utilisation but you must identify and manage the limiting resource.
Leads-to
Throughput
Packet switching increases aggregate throughput by statistical multiplexing—many flows share links. Throughput is the metric; packetisation and routing are the mechanisms. Congestion control keeps throughput high without collapse.
Section 8
One Key Quote
— Paul Baran, On Distributed Communications (1964)"Message blocks could be sent along different paths and reassembled at the destination." Packet switching trades guaranteed path and latency for flexibility and multiplexing—the foundation of scalable networks.
Section 11
Summary & Further Reading
Packet switching sends data in discrete, routable units over shared links, enabling statistical multiplexing and scalable throughput. Variable latency and congestion require buffering and control. Use it when building systems that multiplex work or data over shared capacity.
01
Book
Standard treatment of packet switching, routing, and congestion control.
02
Book
Applies packet-like messaging, queues, and multiplexing to distributed systems.
03