Struct cfxcore::consensus::consensus_inner::confirmation_meter::ConfirmationMeter

pub struct ConfirmationMeter { /* private fields */ }

ConfirmationMeter computes an approximate local view confirmation risk given the current blockchain state. Local view means that the meter assumes a potential block propagation delay and assumes a worst case scenario of what this delay could do.

The meter serves two purposes. First, it allows the underlying storage layer to determine whether it is relatively safe to discard previous snapshots. Snapshot consumes a lot of disk space and it is ideal to discard old ones. Second, it enables the consensus layer to provide an interface to query the confirmation status of a block/transaction.

Implementations

impl ConfirmationMeter

pub fn new() -> Self

pub fn clear(&self)

pub fn update_total_weight_delta_heartbeat(&self)

This is the function that should be invoked every 2 * BLOCK_PROPAGATION_DELAY by the synchronization layer to measure the weight of generated blocks in 2d

pub fn aggregate_total_weight_in_past(&self, weight: i128)

The ConsensusGraph calls this function for every inserted and activated block to accumulate the total weight value

pub fn reset_for_checkpoint(&self, total_weight: i128, stable_height: u64)

The ConsensusGraph invokes this function when making a checkpoint. The confirmation meter needs to aware of the genesis change and make adjustment accordingly.

pub fn get_confirmed_epoch_num(&self) -> u64

pub fn confirmation_risk_by_hash( &self, g_inner: &ConsensusGraphInner, hash: H256 ) -> Option<f64>

Query the confirmation hash of a specific block.

pub fn update_confirmation_risks(&self, g_inner: &ConsensusGraphInner)

ConsensusGraphInner invokes this function to recompute confirmation risk of all epochs periodically

pub fn is_adaptive_possible( &self, g_inner: &ConsensusGraphInner, me: usize ) -> bool

This is an expensive function to check whether the current tree graph will generate adaptive block under me in future. This function is used by Conflux to determine when we will remove old snapshots. If this is true, we will avoid remove snapshots from the storage layer.