Struct DhtRoutingOptimizer 

pub struct DhtRoutingOptimizer {
    pub stale_threshold_secs: u64,
    pub target_bucket_fill: usize,
    pub k_bucket_capacity: usize,
}

Analyzes Kademlia routing table snapshots and emits remediation suggestions.

Example

use ipfrs_network::dht_optimizer::{DhtRoutingOptimizer, RoutingEntry};

let optimizer = DhtRoutingOptimizer::new(600, 8);
let entries = vec![
    RoutingEntry::new("peer-1", 5, 1_700_000_000, 12.0, true),
    RoutingEntry::new("peer-2", 5, 1_700_000_010, 18.5, true),
];
let report = optimizer.optimize(&entries, 1_700_000_600);
assert!(!report.has_issues());

Fields

stale_threshold_secs: u64

Seconds of inactivity before a peer is considered stale.

target_bucket_fill: usize

Target number of responsive peers per bucket.

k_bucket_capacity: usize

Maximum number of entries allowed per bucket (Kademlia k).

Implementations

impl DhtRoutingOptimizer

pub fn new(stale_threshold_secs: u64, target_bucket_fill: usize) -> Self

Creates a new optimizer with the given thresholds.

Arguments

• stale_threshold_secs – seconds without a response before a peer is labelled stale (default: DEFAULT_STALE_THRESHOLD_SECS).
• target_bucket_fill – desired number of responsive entries per bucket (default: DEFAULT_TARGET_BUCKET_FILL).

pub fn analyze_bucket( &self, entries: &[RoutingEntry], now_secs: u64, ) -> BucketAnalysis

Analyzes a slice of entries that all belong to the same bucket.

The bucket_index field of the returned BucketAnalysis is taken from the first entry; callers should ensure all entries share the same bucket index (as guaranteed by optimize).

Health classification order

1. Saturated — entry_count > k_bucket_capacity
2. Stale — any non-responsive entries present
3. Sparse — responsive_count < target_bucket_fill
4. Healthy — otherwise

pub fn optimize( &self, all_entries: &[RoutingEntry], now_secs: u64, ) -> OptimizationReport

Analyzes all routing entries and returns a full OptimizationReport.

Algorithm

1. Group entries by bucket_index.
2. Run analyze_bucket on each group.
3. For each bucket produce recommendations:
   • Stale entries: emit PingPeer (or EvictPeer if age > 2× stale_threshold_secs), capped at MAX_PINGS_PER_BUCKET per bucket.
   • Sparse bucket: emit RefreshBucket.
4. Collate bucket-health counts and return the report.

pub fn top_latency_peers<'a>( &self, entries: &'a [RoutingEntry], n: usize, ) -> Vec<&'a RoutingEntry>

Returns the n responsive peers with the highest latency, sorted descending. These are candidates for replacement by lower-latency peers discovered during bucket refresh.

pub fn coverage_score(&self, entries: &[RoutingEntry]) -> f64

Returns the fraction of the 256 possible Kademlia bucket indices that have at least one responsive entry.

Returns 0.0 for an empty entry list.

Trait Implementations

impl Clone for DhtRoutingOptimizer

fn clone(&self) -> DhtRoutingOptimizer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for DhtRoutingOptimizer

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for DhtRoutingOptimizer

fn default() -> Self

Returns the “default value” for a type.