Consistent Hashing Tests

Purpose

This laboratory work demonstrates the implementation of a consistent hashing algorithm in Go. The project includes:

• A custom implementation of consistent hashing using CRC32 hashing.
• Comparison tests with two popular Go libraries: github.com/ArchishmanSengupta/consistent-hashing and github.com/buraksezer/consistent.
• Benchmarking to measure performance.

The goal is to understand how consistent hashing works, its benefits for distributed systems, and how different implementations compare in terms of load distribution and performance.

Implementation

The main implementation is in main.go, which defines the ConsistentHash struct with methods:

• NewConsistentHash: Creates a new instance with configurable replicas and hash function.
• Add: Adds nodes to the hash ring with virtual replicas.
• Get: Retrieves the node responsible for a given key.

Testing

Tests in main_test.go include:

• Basic functionality test for the custom implementation.
• Comparison with ArchishmanSengupta's library.
• Comparison with Buraksezer's library, including node removal scenarios.
• Load distribution analysis.

Benchmarking

Benchmarks compare the performance of the custom implementation and ArchishmanSengupta's library.

Dependencies

• Go 1.24.3
• github.com/ArchishmanSengupta/consistent-hashing v1.0.2
• github.com/cespare/xxhash v1.1.0
• github.com/buraksezer/consistent v0.10.0

Running the Tests

Run go test to execute the tests and benchmarks.

Run go test -bench=. to run benchmarks.