Self-Claimed Hardware Specification

Each Arx node in the Arcium Network must declare its hardware specification when it is created. Referred to as the "Self-Claimed Hardware Specification", it is defined as the number of Computation Units per unit of time that the node can handle at maximum load. The Arx node is incentivized to make an honest claim, since if the claim is an underestimate then the node will receive less computational work (and therefore fewer rewards), and if the claim is an overestimate then the node could be slashed for non-participation if it is too slow to perform a large Computation (or, for example, crashes when attempting to perform a Computation that's too large for its available hardware).

Arx nodes may freely improve their hardware specifications after launching. However decreasing a node's declared hardware specification after launch may only be done if the node is either (a) not in any Clusters currently, or (b) willing to pay for the migration (from their self-delegated stake) of any of the Clusters that they are in where the lowering of their hardware specification decreases the overall performance of the Cluster such that the new overall performance falls below the Cluster's defined required maximum computational load (see the Cluster Formation section for more details on the Computation Customer's definition of this limit).