Projects sharing features with Mooncake

Sglang is a high-performance inference engine and serving system designed for large language and multimodal models. It provides a programmable interface for orchestrating complex generation workflows, enabling developers to coordinate multi-turn dialogues, tool invocations, and reasoning chains through a domain-specific language. The platform is built to support production-scale deployments, offering an OpenAI-compatible API that allows for integration with existing application ecosystems. The system distinguishes itself through a disaggregated architecture that separates compute-intensive pr

Hazelcast is a distributed data platform that combines an in-memory data grid with a stream processing engine to support real-time analytics and event-driven applications. It functions as a partitioned, distributed key-value store that replicates data across cluster nodes to provide low-latency access and high availability. The platform also serves as a distributed SQL query engine, allowing users to execute standard SQL statements against both in-memory datasets and external data sources. What distinguishes Hazelcast is its use of a distributed consensus subsystem to maintain strongly consis

FastDeploy is a high-performance deployment framework for large language models, vision models, and multimodal models. It provides the infrastructure to launch model services that process combined image, video, and text inputs, exposing these capabilities through a standardized, OpenAI-compatible API for chat and text completions. The project distinguishes itself through advanced inference pipeline engineering and GPU optimization. It employs speculative decoding, tensor parallelism, and a disaggregated execution model that separates prefill and decode phases across different hardware resourc

Memcached is a high-performance, distributed, in-memory key-value storage and request routing engine. It functions as a volatile data store designed to accelerate dynamic applications by caching objects in RAM, thereby reducing backend database load and providing sub-millisecond response times. The system utilizes a specialized architecture that organizes memory into fixed-size slabs to minimize fragmentation and maximize throughput for high-concurrency workloads. The project distinguishes itself through a multi-threaded, lock-friendly design that scales across CPU cores and supports complex

Olric is a distributed data grid and in-memory key-value store that partitions and replicates data across a cluster of servers. It serves as a shared memory system for managing distributed maps, performing atomic operations, and acting as an in-memory data cache. The system provides a distributed locking mechanism for concurrency control and a pub-sub messaging system that broadcasts and routes messages over named channels across the cluster. The platform covers wide-ranging capabilities including cluster management and orchestration, data replication with configurable quorums, and automated

NCCL is a high-performance communication library and distributed GPU computing framework designed for executing collective and point-to-point data exchanges across multiple GPUs in single or multi-node systems. It serves as an RDMA GPU transport layer and memory orchestrator, facilitating high-bandwidth synchronization of data and model gradients for distributed GPU training and inference. The library is distinguished by its ability to execute communication primitives directly from GPU kernels, removing the host CPU from the critical path. It utilizes topology-aware path selection to optimize

node-lru-cache is a key-value store for Node.js that implements a memory-limited cache. It prevents memory exhaustion by capping the total number of items or total byte size, automatically evicting the least-recently-used items to maintain these limits. The project provides specialized caching patterns, including time-to-live durations to ensure data freshness and a stale-while-revalidate mechanism that serves expired content while asynchronously fetching updates in the background. It also includes diagnostic tools for monitoring cache hits, misses, and eviction metrics through operational st

Redis is a high-performance in-memory key-value store that functions as a distributed cache, message broker, and NoSQL database. It provides sub-millisecond read and write access to data stored in RAM and can operate as a vector database for indexing high-dimensional embeddings. The system supports a wide range of data storage and synchronization primitives, including the management of strings, hashes, lists, sets, and JSON documents. It enables real-time data operations through atomic transactions, hybrid persistence using snapshots and append-only logs, and high-availability configurations

This project is a collection of reference implementations and practical guides for building enterprise Java applications using the Spring Boot framework. It serves as a backend project gallery and implementation guide, providing a set of architectures for common server patterns. The repository distinguishes itself through a focus on distributed system design, offering examples for global unique identifier generation, distributed caching, and full-text search. It also includes templates and examples for creating custom Spring Boot starters to encapsulate shared dependencies and configurations

FoundationDB is an ACID-compliant distributed transactional key-value store. It functions as a scalable database engine that ensures strict serializability and data consistency across a cluster of servers using a shared-nothing architecture. The system is distinguished by its multi-region replication capabilities, allowing data to be synchronized across different datacenters for high availability and disaster recovery. It utilizes optimistic concurrency control to manage distributed transactions and employs a majority-based coordination system to maintain cluster state. The platform provides

Twemproxy is a lightweight proxy that routes and distributes requests across multiple Redis and Memcached backend servers. It functions as a protocol translation gateway and distributed cache shard manager, partitioning data across clusters to balance load and storage capacity. The system acts as a high-availability cache orchestrator, employing health monitoring and automatic server ejection to maintain continuous access to cached data. It integrates with sentinels for dynamic master and replica discovery and utilizes consistent hashing and tag-based key grouping to manage data distribution

KeyDB is a multithreaded in-memory key-value store and distributed cache. It functions as a NoSQL database utilizing multi-version concurrency control to execute non-blocking queries and scans. The project is a multithreaded fork of Redis that maintains protocol compatibility while utilizing a multithreaded architecture to scale across multi-core hardware. It distinguishes itself with flash-tiered storage, allowing the system to offload data from primary RAM to SSD or flash storage to increase total capacity. The system supports high availability through active-active mesh replication and mu

Garnet is a multi-threaded in-memory database and distributed key-value store. It functions as a high-performance remote cache store that implements the RESP wire protocol to maintain compatibility with existing Redis clients and libraries. The project is distinguished by a shared-memory architecture that enables parallel request processing across multiple cores for sub-millisecond latency. It features a tiered storage system that automatically offloads colder data from system memory to SSD or cloud storage layers, and includes a specialized vector search database for high-dimensional similar

KServe is a Kubernetes-native platform for deploying and serving machine learning models as scalable inference services. It supports both generative AI models, including large language models, and traditional predictive models from frameworks such as TensorFlow, PyTorch, Scikit-Learn, XGBoost, and ONNX. The platform manages the full lifecycle of model deployments, including revision tracking, canary rollouts, A/B testing, and automatic rollbacks, and provides serverless scale-to-zero capabilities for cost-efficient resource management. KServe distinguishes itself through a standardized infere

CRMEB is a comprehensive e-commerce platform built on ThinkPHP 6, designed as a headless system that delivers standardized APIs to various frontend clients. It provides a unified backend to synchronize product catalogs, orders, and customer data across web browsers, mobile applications, and mini-programs. The platform supports diverse commerce models, including multi-vendor marketplaces where independent merchants manage their own stores, centralized chain store networks, and social commerce frameworks featuring affiliate distribution and community group buying. It also integrates specialized

LeetCode-Swift is a collection of algorithm solutions written in Swift, designed for coding interview preparation. Each solution is implemented as a self-contained function with no external dependencies, making it easy to run and test. The repository organizes solutions by topic and company, and every file includes time and space complexity annotations, allowing quick evaluation of algorithmic efficiency. What sets this repository apart is its flat file structure and the way solutions are tagged with the companies that asked them in interviews, enabling targeted practice. All code resides in

HanekeSwift is a generic caching library for iOS and a specialized image caching framework. It provides a multi-level system that stores arbitrary data types in memory and on persistent disk storage to reduce network requests. The project features a specialized image handler that manages asynchronous loading, resizing, and disk storage for user interface components. It includes a background retrieval system that fetches remote content and automatically populates local caches. The library covers key-value data storage with sequential fallbacks, where it checks memory, then disk, and finally r

Aibrix is an inference orchestrator designed for scaling, routing, and managing the deployment of large language models across distributed vLLM clusters. It serves as a centralized gateway for load balancing and routing traffic to specific model replicas and versions. The system manages resource efficiency through a GPU cluster autoscaler that adjusts compute instance counts based on real-time request volume. It further optimizes operations by mixing different accelerator types within a single cluster and utilizing a model adapter orchestrator to deploy lightweight parameter adapters on share

This project is a comprehensive performance programming guide and reference for the Go language, focusing on runtime efficiency and memory optimization. It provides a collection of patterns and techniques designed to increase execution speed by reducing garbage collection overhead and optimizing memory usage. The resource distinguishes itself through detailed reference implementations for memory optimization, such as escape analysis, object pooling, and structure memory alignment. It offers specific strategies for reducing binary size and improving CPU cache efficiency through structure memor

TiKV is a cloud-native distributed transactional key-value store and storage engine. It provides a distributed database designed for horizontal scalability and strong consistency across a cluster of physical nodes. The system uses a Raft-based consensus mechanism to maintain data availability and state synchronization. It ensures ACID compliance for distributed transactions through a two-phase commit workflow and manages data distribution via multi-Raft sharding. The engine handles massive datasets using automated range splitting and cluster load balancing to distribute data across different

SLIME is a distributed reinforcement learning framework for large language model post-training that bridges Megatron training with SGLang inference servers. It orchestrates scalable RL loops across GPU clusters, decoupling training and inference into independent processes that communicate over HTTP and NCCL for independent scaling and fault tolerance. The system supports multi-agent reinforcement learning workflows with parallel agent instances, customizable rollout strategies, and personalized agent serving that improves models from prior conversations without disrupting API serving. The fra

Groupcache is a distributed caching library designed to coordinate data retrieval and storage across a cluster of nodes. It functions as a peer-to-peer data store that uses consistent hashing to assign specific keys to canonical owners, ensuring that cached items remain predictable and accessible throughout the network. The system distinguishes itself through a request coalescing engine that merges concurrent requests for the same missing key into a single upstream fetch. This mechanism prevents redundant backend load by ensuring that only one process retrieves the required data while sharing

Kvrocks is a disk-based NoSQL database and distributed key-value store that leverages the RocksDB storage engine to persist large datasets to physical disk. It is designed to be a Redis-compatible database, utilizing the standard Redis communication protocol to ensure interoperability with existing client libraries and tools. The project distinguishes itself by combining a disk-persistent storage model with advanced retrieval capabilities, including vector search for k-nearest neighbor queries, full-text search indexing, and geospatial query execution. It supports distributed clustering with

TiKV is a distributed transactional key-value store designed for horizontal scalability and high availability. It functions as a storage engine that maintains massive datasets across a cluster of physical nodes, ensuring that information remains accessible and consistent even when individual hardware components fail. The system utilizes a consensus-based replication model to synchronize data across nodes, ensuring that all replicas agree on the order of operations. It manages data distribution through a sharding mechanism that partitions large datasets into smaller groups, each governed by in

LightLLM is a high-performance serving framework for deploying and executing large language models. It functions as a multi-GPU inference engine and server capable of handling dense architectures, mixture-of-experts designs, and multimodal models that process both text and images. The system is distinguished by its specialized support for Mixture-of-Experts models using expert parallelism and fused kernels. It implements structured text generation through deterministic state machines and pushdown automata to enforce precise output formats. To optimize throughput, the framework employs specula

Dynomite is a distributed data sharding layer and key-value storage engine proxy. It functions as a distribution layer that shards and replicates data across multiple nodes, transforming single-server datastores into scalable peer-to-peer systems. The system acts as a multi-datacenter data replicator, synchronizing data between different geographical locations to ensure resilience and high availability during site failures. It manages the distribution of key-value data to enable linear datastore scaling and redundant storage. The project provides capabilities for storage engine sharding and