Comparison
Blast radiusvsCell-based architecture
Blast radius
you ask how much of the product breaks if this one component dies, and design so the answer is not "all of it".
The scope of impact when a given component fails. Reducing it is the point of cells, shards, bulkheads and per-tenant isolation. It is also the right frame for reviewing a change: not "will this break" but "what breaks with it".
Full entry →Cell-based architecture
customers are divided across eight complete copies of the stack, so a bad deploy or a poison request takes out an eighth of them.
Partitioning the whole system into independent, complete instances and assigning each customer to one. It converts most total outages into partial ones and gives deployments a natural containment unit. The costs are operational multiplication — everything you do, you do eight times — and routing, since something has to map a customer to a cell and that router is now the shared dependency.
Full entry →