Chionanthus

Chionanthus

Optimizes networking and prevents common connection issues

by
86 Downloads
fabricoptimization
Rent Server with this Mod

About this Mod

Chionanthus

Available on Fabric
Available on Modrinth
Available on GitHub

Overview

Chionanthus is a high-performance network optimization mod for Fabric that dramatically reduces bandwidth overhead and improves connection efficiency. Through intelligent packet batching, compression pooling, and adaptive flush strategies, Chionanthus optimizes Minecraft's network layer without compromising compatibility or stability.

Features

Network Layer Optimizations

Intelligent Packet Batching

  • Consolidates multiple packets into single network operations
  • Reduces system call overhead compared to vanilla
  • Three adaptive flush strategies: Adaptive, Tick-Aligned, and Thread-Local

Advanced Compression Handling

  • Thread-local object pooling eliminates repeated Deflater/Inflater allocation
  • Buffer pooling reduces byte array allocations during compression
  • Configurable compression levels with hot-reload support

Optimized Packet Framing

  • Custom variable-length integer encoding/decoding
  • Reduced overhead in packet length encoding
  • Streamlined frame decoder and prepender implementations

Server List Performance

Dynamic Thread Pool Management

  • Adaptive thread pools that scale with server count
  • Prevents resource waste from vanilla's fixed 50-thread pool
  • Intelligent timeout handling prevents UI freezes on unresponsive servers
  • Proper lifecycle management prevents thread leaks

Performance Monitoring

Built-in Metrics System

  • Tracks packet counts, compression ratios, and flush efficiency
  • Thread-specific flush pattern analysis
  • Optional debug logging at configurable intervals
  • Zero performance impact when disabled

Technical Details

To achieve its optimizations, Chionanthus:

  • Injects custom handlers into Netty pipelines during connection initialization
  • Replaces vanilla compression handlers with pooled, high-performance alternatives
  • Manages thread-local resource pools for lock-free, per-thread optimization
  • Uses atomic primitives for lock-free coordination in flush handlers

Flush Strategies

Adaptive Flush Handler
Dynamically adjusts batching based on current packet rate for optimal performance across varying network conditions.

Tick-Aligned Flush Handler
Batches packets per game tick (50ms intervals) for consistent, predictable network behavior.

Thread-Local Flush Strategy
Applies thread-specific thresholds optimized for render, server, and IO threads.

Thread Safety

All optimizations are designed with thread safety in mind:

  • ThreadLocal pooling provides lock-free per-thread instances
  • Atomic operations eliminate synchronization overhead
  • Proper volatile usage ensures visibility across threads

Configuration

Chionanthus provides extensive configuration options:

Network Settings

  • Maximum packet size: Configure packet size limit for mods and data packs with large data transfers
  • Compression level: Fine-tune compression
  • Flush consolidation: Toggle packet batching on/off
  • Flush strategy: Enable/disable Adaptive, Tick-Aligned, and Thread-Local strategies
  • Base flush threshold: Adjust batching aggressiveness
  • Max flush delay: Set latency cap in milliseconds

Server List Settings

  • Thread pool sizing: Minimum/maximum threads and overhead calculation
  • Thread keep-alive time: Control thread lifecycle
  • Pool recreation threshold: Prevent resource drift
  • Ping timeout: Prevent user interface freezes on unresponsive servers

Debug & Monitoring

  • Performance metrics: Optional detailed logging (performance impact when enabled)
  • Metrics interval: Configure logging frequency
  • Hot-reload: Apply configuration changes without restart (where supported)

Performance Impact

Flush Consolidation: Reduces from 1 syscall per packet to 1 syscall per batch

Compression Pooling: Eliminates allocation overhead by reusing pooled instances

Thread Pool Scaling: Dynamic sizing based on actual server list size vs vanilla's fixed 50 threads

Available Versions

Chionanthus 1.0.0 - Hotfix 1 (?)release
MC 1.19, 1.19.1, 1.19.2fabric
February 27, 2026
Chioanthus 1.0.0release
MC 1.19, 1.19.1, 1.19.2fabric
November 30, 2025

How to Install Chionanthus on Your Server

1

Order Server

Order a Minecraft Java server with at least 3 GB RAM (4 GB recommended).

2

Set fabric Loader

In the panel under "Egg", select the fabric loader and matching Minecraft version (1.19.2).

3

Install Mod

Open the mod browser in the dashboard and search for "Chionanthus". Click "Install" – done! Alternatively, upload the .jar via SFTP to the /mods folder.

Compatibility

Mod Loaders

fabric

Minecraft Versions

1.19.2, 1.19.1, 1.19

Server-side

~ Optional

Recommended RAM

4 GB(min. 3 GB)

Frequently Asked Questions

Chionanthus server crashes on startup – what to do?

Most common cause: wrong fabric version or insufficient RAM. Check the server log (latest.log) for "OutOfMemoryError" or "Mixin" errors. With Mado Hosting: ensure at least 3 GB RAM is allocated and the loader matches the mod version (1.19.2). You can switch loaders with one click in the panel.

Is Chionanthus compatible with fabric?

Chionanthus officially supports fabric for Minecraft 1.19.2, 1.19.1, 1.19. The Mado dashboard automatically detects incompatible loader combinations.

Server lagging with Chionanthus – how to optimize performance?

Recommended RAM: 4 GB (per 8 players). Use /spark profiler to check if Chionanthus consumes the most tick time. Common fixes: reduce server view-distance to 8-10, install "performant" or "starlight" as supplementary mods on Forge. With Mado Hosting, your server runs on NVMe SSDs with dedicated CPU cores for minimal latency.

Rent Modded Server

Install Chionanthus with just one click on your server.

Recommended RAM
4 GBab €8/mo
Min. 3 GB | +1 GB pro 8 Spieler
Create Server Now
1-Click Mod Install
NVMe SSD Storage
DDoS Protection included

Details

License
MIT No Attribution
Server-side
Optional

Supported Versions

1.19.21.19.11.19