Network Stakeholders

The Arcium Network consists of three Network stakeholders who engage with the Network economically to either obtain or provide services. This section serves to detail these roles and the specifics of their respective Network incentives.

Computation Customers

As the Network's buy-side participant type, Computation Customers purchase bespoke confidential computing services from the Arcium Network. To access computing services, a Computation Customer must first create a computation environment (see MXEs), select a Cluster to run their environment on top of, and then define the details of the types of Computations that they plan to have computed. Customers may then order the execution of Computations in accordance with their pre-defined Computation types by paying the corresponding Computation fee which consists of a Network-defined base cost, plus a customer-defined priority fee, enabling customers to have their Computations prioritized for higher fees (see the Pricing section for more). Large computational operations, as well as those which occur on an on-going basis, may be broken down into smaller individual Computations that are then strung together, the output results from one feeding into the parameters of the next and so on.

Arx Operators

Arx Operators are the service providers of the Arcium Network, they run the Arx MPC node software on their physical hardware infrastructure in order to participate in the execution of Computations, earning rewards proportional to the computational jobs they complete. When launching an Arx node, the operator specifies the amount of computational resources their node can provide to the Network, however, in order for the hardware resources to become eligible to do work, the total amount of collateral staked to the Arx node must be proportional to its computational resources (see the Staking section for more). Operators must also specify if their Arx node is equipped with a TEE, enabling participation in Computations where the Computation Customer specifically requires the use of a TEE. In addition to available computational resources and TEE-status, Computation Customers also select Arx nodes (in Clusters) based on their historical reputations (see the Slashing section). Thus, an Arx Operator focuses on designing and operating a maximally reliable service to maintain their reputation in order to receive a continual flow of computational work (reputation is assessed off-chain by Computation Customers based on, for example, historical slashing record, completed Computation statistics, etc.). To further grow their business, an operator can opt to increase the amount of infrastructure resources that they provide to the Network.

Arx Operators have two income streams, the first being rewards resulting from an obligatory self-delegation of stake, and the second being fees charged on 3rd-party delegations to their Arx nodes (see the 3rd-Party Delegators section below). The obligatory minimum self-delegation of stake is used to cover the costs of scheduled migrations if needed.

3rd-Party Delegators

Each 3rd-Party Delegator delegates their stake to an Arx node operating on the Arcium Network in order to receive a pro-rata share of the Arx node's rewards from executing Computations. A rate-based fee, charged by Arx Operators, is deducted from 3rd-Party Delegators' rewards (since Arx Operators assume the actual physical infrastructure provision costs). 3rd-party stake helps an Arx Operator "unlock" their hardware, making it eligible to do work. Furthermore, 3rd-party stake is liable to be slashed in the event of misbehaviour, meaning that 3rd-Party Delegators assume the same collateral risks as Arx Operators.

Arx Operators can optionally disable 3rd-Party Delegations, effectively only permitting their own self-delegation to their Node. For more on this option see the Optionality of 3rd-Party Delegations section.