Number of epochs on Solana during which a validator has been active, meaning it has been producing credits for the epoch.
Percentage growth of the investment (stake account balance) in case rewards are taken out. We estimate the validators' APY for each epoch based on that epoch's duration, cluster's SOL supply, validators' earned credits and stake.
Let:
-
$E$ be epochs per year, -
$I_Y$ be cluster inflation per year, -
$I_{Taper}$ be inflation taper, -
$1 - I_{E}$ be inflation taper per epoch, -
$C_v$ be $v$th validator's commission, -
$S_{Supply}$ be cluster's total supply of SOL, -
$V_{i_{Credits}}$ be credits earned by $i$th validator, -
$V_{i_{Stake}}$ be $i$th validator's active stake, -
$p_v$ be $v$th validator's share of inflation rewards, -
$R_{v{_Y}}$ be inflation rewards generated by the validator in a year, -
$R_{v_{Staker}}$ be inflation rewards shared with the stakers over a year,
then:
$$R_{v{Y}} = \sum{e=0}^{E-1} \frac{I_Y \cdot I_E^e \cdot p_v \cdot S_{Supply}}{E} = \frac{I_Y \cdot p_v \cdot S_{Supply} \cdot (I_E^E - 1)}{E \cdot (I_E - 1)}$$
Epochs' durations, cluster SOL supply, credits earned by validators, total amount of staked SOL - all of these change throughout the year and therefore it is virtually impossible to accurately project the staking yield. Our calculations are to be considered as mere optimistic estimations.
Percentage growth of the investment (stake account balance) over the year given the rewards earned per every epoch are left inside the stake accounts and therefore earn additional rewards. Please note that this is just an estimate (see APR). Let:
-
$APR_v$ be APR of the $v$th validator, -
$E$ be epochs per year,
then:
List of validators (identity keys) that have been manually blacklisted by Marinade team. A validator can get blacklisted for the following reasons:
- Raising its commission at the very end of epochs to steal the rewards from its stakers
- Cheating with credits by vote lagging
Percentage of the staking rewards that will be taken by the validator for the epoch. Marinade's delegation strategy allows validators to go up to 10% commission, and a bonus to Marinade score is applied for validators running with a lower commission. The bonus is a multiplier applied to score that works according to the following rules:
- 10% commission - *1 to the score
- 9% commission - *2 to the score
- 8% commission - *3 to the score
- 7% commission - *4 to the score
- 6% commission and less - *5 to the score
In the Solana network, an epoch has a variable time and corresponds to the time a leader schedule is valid. An epoch is 432 000 slots long, with a target slot time of 400 ms (which equals approximately 2 days), but the target slot time is not always achieved making the epoch lenght variable. You can follow the evolution of the current and previous epochs on Solana explorers or directly on Marinade.
Public key that represents a validator on the Solana network.
Each validator is ranked every epoch by Marinade's delegation strategy, according to the set of criteria it takes into account. Marinade rank represents the relative position of the validator according to Marinade's scoring system.
Ratio, for a given validator, between the stake received by Marinade and the stake received by other users (including the validator itself self-staking).
Numerical label that corresponds to the unique IP of the validator (or node). The public IP address of the validator is used for network communication with the validator. It also hints at the validator's geolocation which is an important decentralization factor.
Geographical location where the validator (or node) is hosted, deduced by its IP using maxmind DB.
Estimation of the stake to receive or to lose from Marinade for a given validator according to its current score and the current state of the pool. Depending on the context in the dashboard, this metric can represent the estimated stake to receive or lose for the next epoch, or the difference between the current stake of a validator and the stake it should receive overall.
Company owning the data center in which a given validator (or node) is hosted, deduced from its public IP.
Ordering of all validators by active stake.
In the Solana network, a skipped slot is a slot where the leader did not produce any block, either because it was offline (delinquent) or because the consensus of validators on-chain followed a different fork. Validators should aim to skip as little slots as possible. The leader schedule assigns slots in batches of 4 to different validators and each slot can produce 1 or 0 block, a slot containing 0 block is considered skipped.
Detailed view of the different stake accounts being delegated to a given validator.
Amount of staked SOL that is delegated to different validators that have the same geographical location. A high concentration of stake in the same city centralizes the network geographically. This metric is taken into account by Marinade's delegation strategy to decentralize stake geographically as much as possible.
Visualization of the "Marinade stake/all active stake" metric.
In the Solana network, the superminority is composed of the largest validators representing, combined, 33.3..% of all the SOL staked. This set of validators has the power to halt the chain if they were to become delinquent at the same time. Marinade's delegation strategy only delegates outside of the superminority. As a note, it is important to remember that any set of validators representing more than 33.3..% of the total SOL staked on the network have the possibility to halt the network, but the superminority represents the lowest number of validators that meet this criteria.
Amount of time, in percentage, during which a given validator has been online. Validators have a duty to strive for the highest uptime for possible and only go offline for upgrades.
Version of the Solana client used by a given validator.
Keypair, represented publicly by a public key that is a unique to a given validator and corresponds to the identity that the validator is using to vote to achieve consensus on-chain.