GS

A plain-language, complete guide to why cache invalidation is considered one of computing’s hardest problems: what a cache really is, what “invalidation” actually means, a short history of the famous joke, why we cache in the first place, the four core reasons the problem is so hard (too many copies,…

A plain-language, complete guide to Content Delivery Networks: what a CDN really is, a short history from the 1990s to today, how it actually works with DNS and Anycast routing, the building blocks (PoPs, edge servers, origin, traffic director), origin pull vs origin push, how to spot a CDN at…

A plain-language, complete guide to throughput: what it really is, a short history of the idea from factories to fibre-optic cables to Kanban boards, how throughput is measured in networking, how it differs from bandwidth and from latency, what affects it, how it shows up in project management, how it…

A plain-language, complete guide to latency: what it really means, a short history of the idea, how latency is measured (with percentiles), how it differs from throughput and bandwidth, the four causes (propagation, transmission, processing, queuing), latency across connection types (fibre, Wi-Fi, mobile, copper, satellite), the different kinds of latency…

A plain-language, complete guide to the difference between latency and throughput: what each one really is, a short history of the distinction, how each is measured, what shapes them, how they interact, the surprising math of utilization and the cliff edge queueing theory predicts, the TCP vs UDP trade-off, where…

A plain-language walkthrough of the classic system-design interview moment. Covers: the big idea, why interviewers love the 100x question, what 100x really means in numbers, the four flavours of growth (more users, more requests, more data, wider geography), the mindset shift from correctness to survival, a reusable five-move loop, and…

A plain-language guide to capacity planning in system design: what it really is, why it matters, the different flavours of growth, the core ingredients (traffic, data, bandwidth), how to estimate traffic and peak load, storage and bandwidth, how to turn numbers into a server count, planning component by component, planning…

A plain-language guide to load balancers: what they really are, why they matter, how they work step by step, the two families of algorithms (static and dynamic) with every common variant, different kinds of traffic to balance, Layer 4 vs Layer 7, types by scope, hardware vs software, health checks…

A plain-language guide to the two most fundamental types of load balancing: static (round robin, weighted round robin, random, source IP hash, URL hash) and dynamic (least connections, weighted least connections, least response time, resource-based). Covers each algorithm in turn, a side-by-side comparison, real performance impact, how to choose between…