memcached

Memcached

Memcached is a high-performance, distributed memory object caching system. It is simpler than Redis. It does arguably one thing: Key-Value caching of strings/objects in RAM.

When to Use

• High-Throughput In-Memory Key-Value Caching: Caching database query results, HTML fragments, and session tokens with sub-millisecond latency.
• Pure Ephemeral Caching: Storing disposable data where eviction under memory pressure is acceptable (LRU eviction).
• Multi-Threaded Multi-Core Scalability: Maximizing cache operations on large multi-core servers using Memcached's native multi-threaded architecture.
• Simple Key-Value Lookups: Caching simple string and serialized objects where advanced data structures (Redis sets/hashes) are unnecessary.

Quick Start

# Telnet interface
set mykey 0 60 4
data
STORED

get mykey
VALUE mykey 0 4
data
END

Core Concepts

Multi-Threaded Event Loop & Slab Allocation

Memcached allocates memory in fixed slabs to prevent operating system memory fragmentation:

Memory Pool (Slab Allocator)
  ├── Slab Class 1 (Chunk Size: 96 Bytes)  -> Stores small keys & tokens
  ├── Slab Class 2 (Chunk Size: 120 Bytes) -> Stores session snippets
  └── Slab Class N (Chunk Size: 1MB)       -> Maximum item size

Consistent Hashing Client-Side Routing

Clusters scale horizontally by having client drivers calculate server destinations using consistent hashing algorithms (Ketama):

// Memcached Consistent Hashing in Node.js
import Memcached from "memcached";

const memcached = new Memcached(
  ["10.0.1.1:11211", "10.0.1.2:11211", "10.0.1.3:11211"],
  {
    retries: 2,
    retry: 10000,
    remove: true, // Automatically remove failed nodes from ring
  },
);

// Set with expiration (seconds)
memcached.set("user:415:profile", JSON.stringify(userData), 3600, (err) => {
  if (err) console.error("Cache set failed:", err);
});

Atomic Binary Protocol Operations (CAS, INCR)

Guarantees thread-safe counters and optimistic concurrency control via Check-and-Set (CAS):