The Runtime Theory
Software ArchitectureIn production

Service Mesh Explained

Recording in progress
#service-mesh#sidecar#istio#networking

Once you have many services, the network between them becomes a product: retries, timeouts, TLS, routing, and per-hop metrics. A service mesh automates all of it by injecting a sidecar next to every instance and routing all traffic through it. This video explains the sidecar model, the control plane that configures proxies, and what the mesh can't give you.

Topics covered:

  • The sidecar: a proxy injected beside each pod or instance, intercepting all traffic
  • The data plane: Envoy-style proxies doing retries, timeouts, and load balancing
  • The control plane: how proxy configuration is generated and distributed
  • mTLS: encrypting and authenticating every service connection
  • Traffic routing: canary and version-based routing without touching app code
  • Observability: metrics, traces, and logs the mesh produces for free
  • The cost: an extra hop per request and proxy overhead
  • When a mesh is premature: client libraries and eBPF

Related articles

More in Software Architecture

In production
software architecture

The Strangler Fig Pattern

How to replace a legacy system without a rewrite — routing by path, strangling features incrementally, and keeping both systems honest.

Details
In production
software architecture

Domain-Driven Design Basics

The core of DDD without the ceremony — ubiquitous language, bounded contexts, aggregates, and where the model boundary actually goes.

Details
In production
software architecture

Feature Flags in Practice

How feature flags work end to end — evaluation at runtime, kill switches, and canary rollouts — plus the operational discipline they demand.

Details
In production
software architecture

The Saga Pattern Explained

How distributed transactions work without distributed locks — saga steps, compensating actions, and the orchestration vs. choreography split.

Details
In production
software architecture

CQRS Explained

Command query responsibility segregation — separate read and write models, why they diverge, and what you actually gain from the split.

Details
In production
software architecture

API Gateway Patterns

What an API gateway actually does — routing, auth, rate limiting, and aggregation — and which patterns matter for your traffic.

Details
In production
software architecture

Hexagonal Architecture, Visualized

Ports and adapters in one diagram — how hexagonal architecture keeps business logic independent of databases, HTTP, and everything external.

Details
In production
software architecture

Event-Driven Architecture Explained

How event-driven systems work — events, topics, and consumers — and what the broker actually guarantees in a distributed system.

Details
In production
software architecture

Microservices vs. Monolith

The honest engineering comparison — deployment boundaries, data ownership, failure domains, and the real costs of splitting a system.

Details

Depth, delivered weekly

One technical dispatch a week — articles and episode notes before they go public.

One technical dispatch per week. No noise.