7 releases (4 breaking)
0.5.0 | Aug 21, 2020 |
---|---|
0.4.1 | Aug 14, 2020 |
0.3.1 | Aug 13, 2020 |
0.2.0 | Aug 11, 2020 |
0.1.0 | Aug 10, 2020 |
#29 in #key-hash
11KB
232 lines
Consistent Hash
Consistent Hashing is a distributed system algorithm for assigning values to partitions in a highly dynamic environment.
Example:
Using the algorithm to LB redis cache. (Crate redis = "0.17.0"
).
While the key must implement the hash trait to define a hashing strategy.
The Value must implement the Evict trait to define rebalancing behavior.
let mut ring: ConsistentHash<String,redis::Client> = match ConsistentHash::new(100).unwrap();
ring.add_node("redis://my-redis-node-1".to_string(), redis::Client::open("redis://my-redis-node-1")?).unwrap();
ring.add_node("redis://my-redis-node-2".to_string(), redis::Client::open("redis://my-redis-node-2")?).unwrap();
ring.add_node("redis://my-redis-node-3".to_string(), redis::Client::open("redis://my-redis-node-3")?).unwrap();
ring.add_node("redis://my-redis-node-4".to_string(), redis::Client::open("redis://my-redis-node-4")?).unwrap();
let client = match ring.get_node(String::from("some-key")){
Some(node) => node,
None => panic!("Not found!"),
};
let mut con = client.get_connection()?;
con.get("some-key");
Dependencies
~1.4–2MB
~37K SLoC