πŸ¦€ Built with Rust Open Source

The Fastest Tag-Based Cache Ever Built

Surgical cache invalidation through intelligent tagging.
Atomic operations at wire speed for distributed systems.
1M+ ops/sec throughput with <0.8ms P95 latency.

Performance Metrics

1M+ ops/sec

TCP Throughput

<0.8 ms

P95 Latency

<40 ms

Cold Start

<4 %

Memory Overhead

Why Development Teams Choose TagCache

Built from the ground up for modern cloud-native applications and distributed microservices architectures

Blazing Fast Cache Performance

Lock-minimized architecture with sharded maps and async IO, optimized for multi-core systems and maximum throughput in distributed cache clusters.

Smart Cache Tag Invalidation

Attach multiple tags to cache entries for surgical purges. Invalidate by user, tenant, or any custom dimension instantly across distributed backend systems.

Atomic Operations

Thread-safe INCR, DECR, and ADD operations perfect for rate limiting, counters, and distributed coordination.

Dual Protocol Support

Choose HTTP with JSON for simplicity or binary TCP protocol for maximum performance and minimal overhead.

Full Observability

Beautiful React-based admin dashboard with real-time metrics, structured logging, health checks, and comprehensive performance monitoring for complete cache visibility.

Deploy Anywhere

Docker images, Kubernetes manifests, systemd units, and cloud templates for seamless deployment.

Rust Powered

Memory-safe, zero-cost abstractions, and fearless concurrency for unmatched reliability and performance.

Multi-Shard Concurrency

DashMap-powered sharding eliminates lock contention, enabling true parallel access across CPU cores for maximum throughput.

Get Started in Seconds

Choose your preferred installation method

1
2
3
4
5
6
# Install via Homebrew
brew tap aminshamim/tap
brew install tagcache

# Start the server
tagcache server
1
2
3
4
5
6
7
# Run with Docker
docker run -d --name tagcache \
  -p 8080:8080 -p 1984:1984 \
  ghcr.io/aminshamim/tagcache:latest

# Check status
docker logs tagcache
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
# Linux x86_64
curl -L -o tagcache-linux-x86_64.tar.gz \
  https://github.com/aminshamim/tagcache/releases/download/v1.0.8/tagcache-linux-x86_64.tar.gz
tar xzf tagcache-linux-x86_64.tar.gz
sudo cp tagcache /usr/local/bin/

# macOS Intel
curl -L -o tagcache-macos-x86_64.tar.gz \
  https://github.com/aminshamim/tagcache/releases/download/v1.0.8/tagcache-macos-x86_64.tar.gz
tar xzf tagcache-macos-x86_64.tar.gz
sudo cp tagcache /usr/local/bin/

# macOS Apple Silicon
curl -L -o tagcache-macos-arm64.tar.gz \
  https://github.com/aminshamim/tagcache/releases/download/v1.0.8/tagcache-macos-arm64.tar.gz
tar xzf tagcache-macos-arm64.tar.gz
sudo cp tagcache /usr/local/bin/
1
2
3
4
5
6
7
8
9
# Debian/Ubuntu (APT)
curl -L -o tagcache.deb \
  https://github.com/aminshamim/tagcache/releases/download/v1.0.8/tagcache_1.0.8_amd64.deb
sudo dpkg -i tagcache.deb

# RHEL/CentOS/Fedora (RPM)
curl -L -o tagcache.rpm \
  https://github.com/aminshamim/tagcache/releases/download/v1.0.8/tagcache-1.0.8-1.x86_64.rpm
sudo rpm -ivh tagcache.rpm
1
2
3
4
5
6
7
8
# From source with Cargo
cargo install --git https://github.com/aminshamim/tagcache.git

# Or clone and build
git clone https://github.com/aminshamim/tagcache.git
cd tagcache
cargo build --release
sudo cp target/release/tagcache /usr/local/bin/

Simple Yet Powerful API

Tag-based invalidation makes cache management intuitive

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
# Set with tag(s)
curl -X POST :8080/set \
  -H 'Content-Type: application/json' \
  -d '{"key":"user:42:profile","value":{"name":"Ada"},"tags":["user:42","profile"]}'

# Invalidate all profile data for user 42
curl -X POST :8080/tag/delete \
  -H 'Content-Type: application/json' \
  -d '{"tag":"user:42"}'

# Atomic increment (rate limit / counter)
curl -X POST :8080/increment -d '{"key":"ratelimit:login","delta":1}'
</div>

Integrates With Your Technology Stack

Native cache client support for all major programming languages and cloud platforms

Rust

First-class runtime and native async client library

View API β†’

Kubernetes

Helm charts and horizontal scaling guides

Deploy β†’

Java & Spring

Spring Boot integration with connection pooling

Examples β†’

.NET & C#

Async HttpClient patterns for microservices

Examples β†’

Current Limitations & Roadmap

Transparent about current constraints and exciting features coming soon

In-Memory Only

Currently no persistence to disk. Cache data is lost on restart. Future: Optional persistence with configurable write-ahead logging.

Single Node Architecture

No built-in replication or clustering support. Roadmap: Consistent hashing with peer discovery and automatic failover.

No Compression

Binary protocol lacks compression layer for large payloads. Planned: Optional compression with configurable algorithms (LZ4, Zstd).

Unbounded Tag Cardinality

No automatic limits on unique tag counts. Monitor memory usage with high tag diversity. Future: Configurable tag limits and cleanup policies.

πŸ—ΊοΈ Roadmap Priority: Clustering and replication are top priorities for v2.0, followed by persistence options and compression enhancements.

Ready to accelerate your stack?

Adopt TagCache as your unified low-latency caching layer with surgical invalidation and observability from day one.

Need help? Search the docs, check our GitHub discussions, or open an issue.