off-heap

A local reactive cache for Java and Android. Now, it supports heap memory、off-heap memory and disk cache.

High-performance embedded off-heap storage engine for Java, built on memory-mapped files. Map/List/Set/Queue beyond the JVM heap, plus an AI memory layer with hybrid vector + BM25 search.

Single-threaded off-heap object store

Read-only primitive Java arrays backed by Direct Buffers and indexed using 64-bit indexes

A high-performance off-heap memory pool for Java applications.

Native ByteBuffers for Java

Leaderless, off-heap, zero-GC tri-temporal δ-CRDT database engine. Post-quantum transport. Source-available BSL-1.1. Research preview.

Making DuckDB's capabilities available to the JVM efficiently. Includes a DuckDB-backed CQEngine persistence plugin: 7x less memory, and joins across collections.

Ultra-low latency, zero-copy IPC for Java & Python using memory-mapped files.

A fast, log structured key-value store.

Zero-allocation text classification in pure Go. Byte-level n-gram embeddings, OVA logistic regression, Platt scaling, conformal prediction. Sub-millisecond, GC-free.

High-performance off-heap cache for Java 25+ using the Foreign Function & Memory API. No Unsafe. Built for large, low-latency in-process caches.

From-scratch distributed messaging system — append-only log, binary TCP over NIO, MurmurHash3 partitioning, and hand-rolled Raft consensus with leader election, log replication, and integration tests. Phase 5: off-heap storage + JMH benchmarks on bare-metal Linux.

Off-heap matrices for JBLAS library (Rust -> C++/JNI -> Java)

Researching tail-latency trade-offs in Java caching: Caffeine on-heap vs Chronicle Map off-heap using reproducible JMH benchmarks.

Classical, SAT, modal, temporal, and fuzzy logic — a complete reasoning engine in pure Go. Off-heap, race-clean, zero GC pressure.

Java off-heap alloction using unsafe.

Zero-allocation Binary Decision Diagrams (OBDD) for Go — full Buddy parity with off-heap memory, level indirection, distinct types, and modal logic integration.

[Completed] Ultra-low latency data pipeline & real-time monitoring dashboard using Rust, Java 25 (FFM/Vector API), and Oracle DB.

Zero-infrastructure off-heap cache for AI workloads. One annotation in Java, one decorator in Python. No Redis, no Docker.