Difference between revisions of "Node Provider Remuneration"

From Internet Computer Wiki
Jump to: navigation, search
Line 57: Line 57:
 
|}
 
|}
  
After the first nodes were up and running, the decision was made to increase the storage capacity of all nodes to support larger subnet states. This required additional hardware and operational expenses. Once the additional storage is usable by the IC, we propose to update the Gen-1 remuneration model to cover all expenses. This will be detailed in a separate proposal.
+
After the first nodes were up and running, the decision was made to increase the storage capacity of all nodes to support larger subnet states. This required additional hardware and operational expenses. Once the additional storage is usable by the IC, it is proposed to update the Gen-1 remuneration model to cover all expenses. This will be detailed in a separate proposal.
  
 
= Version 2 Remuneration Model =
 
= Version 2 Remuneration Model =

Revision as of 14:28, 27 February 2023

Node providers receive rewards (remuneration) for operating node machines that run the IC network. The single source of truth for node provider rewards is the NNS, where changes can only be made through NNS proposals adopted by the IC community.

This page summarizes the current node provider rewards and serves to discuss proposals for future reward models. The NNS distributes rewards depending on

  • the node hardware version

  • geography

taking into account the fact that different hardware versions have different capex and opex that vary with the geography. The prices listed for hardware are based on quotes received from different vendors. See Node Provider Hardware for more details on validated hardware configurations. The calculations assume an amortization of capex over 4 years. Operational expenses are based on numbers shared by current node providers. The actual hardware and operational expenses may differ depending on negotiation of individual NPs with their vendors. All prices are in XDR.

Remuneration models.png

The above table clarifies the relationship between the different Hardware types and Remuneration models. Gen1 Hardware has been used by Node Providers to set up nodes during Genesis launch. For these Node Providers, what is called V1 (Version 1) Remuneration model applies. The V1 remuneration model is described below. As work is being done to upgrade the Gen1 hardware with additional storage, the V1 remuneration will need to be updated in the future. This updated remuneration, which is called V1.1 (Version 1.1) remuneration, will be described in a separate wiki page once a proposal for this remuneration has been worked out in detail.

For the further growth of the IC network, a Gen2 HW specification has been defined. The new Hardware specification is generic (instead of vendor specific) and supports VM memory encryption and attestation which will be needed in future features on the IC. Since the HW specification is different, a new version of the remuneration, which is called V2 (Version 2) remuneration model is defined, which is described below. In the future, a remuneration will be proposed that is fully decentralized and includes both automated incentives, rewards and penalties. For clarity, this remuneration will be called V3 (Version 3) remuneration.

Version 1 Remuneration Model (Genesis)

The initial remuneration model included a profit margin of 150%, accounting for the fact that early NPs took on a significant risk when they decided to invest into hardware and operations prior to Genesis. In other words, the rewards distributed were 2.5 times the expected expenses. The following table provides an overview of expenses and rewards paid, broken down by geography. Hardware expenses include shipping and installation charges.

HW OpEx over 4 years Total over 4 years Incl. profit margin Reward per month
USA 8676 8090 16766 41915 873
US - FL/GA/CA 8676 12200 20876 52190 1087
EU 8089 12781 20870 52175 1087
Asia 7882 15389 23271 58178 1212

After the first nodes were up and running, the decision was made to increase the storage capacity of all nodes to support larger subnet states. This required additional hardware and operational expenses. Once the additional storage is usable by the IC, it is proposed to update the Gen-1 remuneration model to cover all expenses. This will be detailed in a separate proposal.

Version 2 Remuneration Model

This section presents a proposal for a new remuneration model. Based on the feedback from NPs and the community on the initial V2 remuneration proposal, discussed in this forum post, the following updates are proposed:

  • Higher rewards for the first nodes of a new NP in order to attract more NPs in an effort to improve ownership decentralization.
  • More refined rewards for nodes in new geographies, like South America, Africa, Asia and Australia, to stimulate further geographical decentralization.

The proposal introduces a node reward model parameterized by:

  • Geography multiplier: This multiplier will be lower, namely 2, for regions with many nodes (e.g. Europe and North America), and higher, namely 3, for regions where there are currently limited nodes present (such as Africa and South America)
  • Reduction coefficient (r): The node reward of the n-th node of a Node Provider is multiplied by r ^ (n-1). The reduction coefficient r is dependent on the geography of the Node Provider. As a result, the first node of a Node Provider gets attractive rewards but it is increasingly less attractive to add additional nodes.

The rewards are furthermore dependent on estimated capital and operational expenses that vary based on geographies. A table with the concrete numbers follows below.

In summary, for a geography g, let

  • mult(g) be the geography multiplier
  • capex(g) the capital expenses to acquire a gen 2 node in g in XDR
  • opex(g) the operational expenses over 4 years to run a gen 2 node in g (in XDR)
  • r(g) be the reduction coefficient

The monthly reward for the n-th node of a Node Provider in geography g are defined as follows:

reward(g, n) = ( capex(g)+opex(g) ) * mult(g) * r(g) ^ (n-1) / (4 * 12)

The capital and operational expenses over 4 years are added, multiplied by the geography multiplier, multiplied by the reduction coefficient, and divided by 4 years times 12 months. As a result, rewards for nodes in new geographies and for Node Provider with fewer nodes are higher. Thereby, a geographical and ownership decentralization is incentivized. The following table shows the geography-dependent values and the monthly reward for the first node onboarded.

Geography Capex: Gen2 HW Opex Multiplier Monthly reward for 1st node Reduction coefficient r
USA 19189 11845 2 1294 0.7
US - FL/GA/CA 19189 17842 2 1542 0.7
EU 18283 18713 2 1542 0.95
Asia Singapore 17977 22531 2 1688 0.7
Asia non Singapore 17977 22531 3 2532 0.98
South Africa 21455 22531 3 2748 0.98

For illustration, the below table shows the calculation of the rewards of the 1st to the 10th node for a Node Provider in South Africa.

reward(south africa, n)

= ( capex(south africa)+opex(south africa) ) * mult(south africa) * r(south africa) ^ (n-1) / (4 * 12)

= ( 21’455 + 22’531) * 3 * 0.98 ^ (n-1) / (4 * 12)

= 2748 * 0.98 ^ (n-1)

N-th node Multiplier (rounded) Monthly remuneration (rounded)
Node 1 0.98 ^ (1-1) =1 2748
Node 2 0.98 ^ (2-1) =0.98 2693
Node 3 0.98 ^ (3-1) =0.96040 2639
Node 4 0.98 ^ (4-1) =0.941192 2586
Node 5 0.98 ^ (5-1) =0.92236816 2534
Node 6 0.98 ^ (6-1) =0.9039207968 2483
Node 7 0.98 ^ (7-1) =0.885842309 2434
Node 8 0.98 ^ (8-1) = 0.8681255332 2385
Node 9 0.98 ^ (9-1) =0.8507630226 2337
Node 10 0.98 ^ (10-1) =0.8337477621 2291

Since the aim is to grow the IC network outside of existing geographies, a remuneration proposal for South America, Australia, as well as other regions is prepared, which will be subsequently added to the reward table. The figures in the above table are in XDR, and are based on existing data points from Node Providers and Hardware vendors.

Profit additional node.jpg

The above figure shows the additional cash flow (investments minus CAPEX and OPEX costs) a NP receives for adding an additional node, for three regions (South Africa, Europe and USA) that can be calculated based on the V2 remuneration model. Note that the calculation does not take into account the time value of the cash flows (i.e. discounted value), but purely the cash flows or rewards and expenses.

What is visible from this graph is that adding additional nodes only generates additional cash flow for a specific number of nodes, in a specific region. For example, in Europe adding up to approximately 15 nodes will generate additional cash flow, whereas for the US region this is only 2 nodes.

Potential new NPs can make a full calculation themselves of the Internal Rate of Return, Payback period and cash flow based on their actual CAPEX and OPEX costs. Of course, these actual CAPEX and OPEX costs may differ from the estimated CAPEX and OPEX costs on which the remuneration is based, depending on the specific contracts the new NP has been able to negotiate with hardware vendors and DC operators.

The V2 remuneration proposal only applies for a limited period in which the team is working on a more elaborate and automated remuneration solution in the NNS, the V3 remuneration. This future remuneration will be based on a reduction coefficient formula that applies a combination of NP, Country, Data Center and city. In addition, the future remuneration model will include penalties in case nodes should be unavailable.