ivertowski 

Core abstractions and interfaces for DotCompute compute acceleration framework. Contains accelerator interfaces, kernel definitions, and memory abstractions.

Core runtime and abstractions for DotCompute compute acceleration framework. Provides kernel execution, accelerator management, and service orchestration.

Package Description

Unified memory management for DotCompute. Provides zero-copy buffers, memory pooling, and cross-device memory transfers.

Plugin system for DotCompute compute acceleration framework. Provides hot-reload capability, plugin discovery, circuit breaker patterns, and fault tolerance infrastructure.

Production-ready NVIDIA CUDA GPU backend for DotCompute. Provides GPU acceleration (21-92x speedup) through CUDA with NVRTC compilation, P2P transfers, Ring Kernels with NCCL support, and unified memory.... More information

Production-ready CPU compute backend for DotCompute. Provides SIMD vectorization (3.7x faster) using AVX2/AVX512/NEON instructions, multi-threaded kernel execution, and Ring Kernel simulation. Benchmarked:... More information

Runtime services and dependency injection integration for DotCompute. Provides kernel execution orchestration, automatic kernel discovery, service registration, and DI container integration with... More information

GPU-accelerated algorithms for DotCompute. Includes FFT, AutoDiff, sparse matrix operations, signal processing, and cryptographic primitives.

Metal backend for DotCompute with GPU acceleration for Apple Silicon (macOS/iOS). Foundation complete with native API bindings, device management, unified memory support, and command buffer execution. MSL... More information

GPU-accelerated LINQ extensions for DotCompute. Transparent GPU execution for LINQ queries with automatic kernel generation, fusion optimization, and Reactive Extensions support.

Attributes and enums for DotCompute kernel definition. This package provides the [Kernel], [RingKernel], and [RingKernelMessage] attributes used to define compute kernels.