Struct ShardedMap 

pub struct ShardedMap<K, V, const B: usize>
where
    K: OHash + Pod + Default + Debug + Send + Ord,
    V: Cmov + Pod + Default + Debug + Send,
    BatchBlock<K, V>: Ord + Send,
{ /* private fields */ }

A sharded hashmap implementation. The map is split across multiple partitions and each partition is a separate hashmap. Queries are resolved in batches, to not leak the number of queries that go to each partition.

Parameters

• K: The type of the keys in the map.
• V: The type of the values in the map.
• P: The number of partitions in the map.
• B: The maximum number of non-distinct keys in any partition in a batch.

Implementations

impl<K, V, const B: usize> ShardedMap<K, V, B>

where K: OHash + Default + Debug + Send + Ord, V: Cmov + Pod + Default + Debug + Send, BatchBlock<K, V>: Ord + Send,

pub fn new(capacity: usize) -> Self

Creates a new ShardedMap with the given number of partitions.

pub fn get_batch_distinct<const N: usize>( &mut self, keys: &[K; N], ) -> [OOption<V>; N]

Reads N values from the map, leaking only N and B, but not any information about the keys (doesn’t leak the number of keys to each partition).

Preconditions

• No repeated keys in the input array.

pub fn insert_batch_distinct<const N: usize>( &mut self, keys: &[K; N], values: &[V; N], )

Inserts a batch of N distinct key-value pairs into the map, distributing them across partitions.

Preconditions

• No repeated keys in the input array.
• All of the inserted keys are not already present in the map.
• There is enough space in the map to insert all N keys.

Trait Implementations

impl<K, V, const B: usize> Debug for ShardedMap<K, V, B>

where K: OHash + Pod + Default + Debug + Send + Ord + Debug, V: Cmov + Pod + Default + Debug + Send + Debug, BatchBlock<K, V>: Ord + Send,

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

Formats the value using the given formatter.