Struct CacheCoordinator 

pub struct CacheCoordinator {
    pub clients: HashMap<NodeId, DistributedCacheClient>,
    pub replication: ReplicationFactor,
}

Cluster-level coordinator.

Tracks all node clients and the replication policy for the cluster. In a real distributed system the coordinator would issue RPCs; here it simulates the routing and quorum decisions.

Fields

clients: HashMap<NodeId, DistributedCacheClient>

Map from NodeId to per-node client.

replication: ReplicationFactor

Cluster-wide replication factor.

Implementations

impl CacheCoordinator

pub fn new(replication: ReplicationFactor) -> Self

Create a new CacheCoordinator with the given replication factor.

pub fn add_client(&mut self, client: DistributedCacheClient)

Register a DistributedCacheClient for its local_node.

pub fn remove_client(&mut self, node_id: NodeId)

Remove the client (and node) identified by node_id.

pub fn primary_node_for(&self, key: &[u8]) -> Option<NodeId>

Determine the primary owner of key according to the first registered client’s ring.

Returns None when no clients are registered.

pub fn replica_nodes_for(&self, key: &[u8], n: usize) -> Vec<NodeId>

Return up to n replica nodes for key according to the first registered client’s ring.

pub fn can_write_quorum(&self, key: &[u8], available_nodes: &[NodeId]) -> bool

Simulate a write operation: determine the owner nodes for key and check whether a quorum can be formed from available_nodes.

pub fn can_read_quorum(&self, key: &[u8], available_nodes: &[NodeId]) -> bool

Simulate a read operation: determine the owner nodes for key and check whether a quorum can be formed from available_nodes.

pub fn node_count(&self) -> usize

Return the number of registered clients/nodes.

Trait Implementations

impl Debug for CacheCoordinator

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

Formats the value using the given formatter.