State Growth and Node Cost Drift

仕組み

Every account, contract, and storage slot a chain has ever created generally stays in the active state, and history accumulates on top of it, so disk, memory, and bandwidth requirements ratchet upward. Fee schedules that underprice storage relative to its permanent cost accelerate this, because writing state is paid for once while it is stored forever. As requirements rise, hobbyist and independent operators drop off, applications shift to a handful of hosted node providers, and the population that can independently check the rules shrinks. That concentration turns into a censorship and outage surface even when the consensus layer is working perfectly, and it also raises the barrier to running the minority clients that protect against implementation bugs.

実際に観測できるもの

Chain size, state size, and their growth rates are published, as are the hardware requirements in official documentation and the practical requirements for archive nodes. Count reachable full nodes and look at their geographic and hosting distribution, and look at how much application traffic reaches the chain through a small number of remote procedure call providers. Watch for repricing proposals, state expiry research, and pruning defaults, which indicate whether the issue is being managed.

先例

Ethereum's 2016 denial-of-service attacks exploited operations priced below their true cost and were answered with a gas repricing and a state-clearing change, an early demonstration that resource pricing and node cost are consensus-relevant.

重要性を左右する要因

Look at state growth rate against falling hardware costs, documented minimum specifications, the count and distribution of independent nodes, and how much of the application layer depends on a few hosted providers.