Data Gravity

ELI5 - The Vibe Check

Data gravity is the idea that once you've got a huge pile of data sitting in one cloud or one region, everything else, your services, your analytics, your new features, ends up getting pulled toward building near it, because moving the data itself is slow and expensive. It's like a black hole: the bigger the pile gets, the harder it is to build anything meaningfully far away from it, which is exactly how a lot of vendor lock-in quietly happens.

Real Talk

Data gravity describes the tendency for applications, services, and analytics workloads to be built and deployed near where large datasets already reside, because network egress cost, transfer latency, and data replication effort make moving that data increasingly expensive as its volume grows. It's a key driver of cloud vendor lock-in and a major factor in multi-region architecture decisions, since teams often choose a region or provider based not on ideal infrastructure but on wherever the bulk of their data already lives.

When You'll Hear This

"We can't just move that service to another provider, data gravity means the egress bill alone would be brutal." / "Data gravity is why the new analytics pipeline got built in the same region as the warehouse, not the ideal one."