Skip to content

Releases: mivertowski/RustCompute

v0.1.1 - AccNet & ProcInt Showcase Applications

04 Dec 15:40
@mivertowski mivertowski
v0.1.1
8de6c1a

Choose a tag to compare

View all tags
v0.1.1 - AccNet & ProcInt Showcase Applications

What's New

New Showcase Applications

AccNet - GPU-Accelerated Accounting Network Analytics

  • Network visualization with force-directed graph layout
  • Fraud detection: circular flows, threshold clustering, Benford's Law violations
  • GAAP compliance checking for accounting rule violations
  • Temporal analysis for seasonality, trends, and behavioral anomalies
  • GPU kernels: Suspense detection, GAAP violation, Benford analysis, PageRank

ProcInt - GPU-Accelerated Process Intelligence

  • DFG (Directly-Follows Graph) mining from event streams
  • Pattern detection: bottlenecks, loops, rework, long-running activities
  • Conformance checking with fitness and precision metrics
  • Timeline view with partial order traces and concurrent activity visualization
  • Multi-sector templates: Healthcare, Manufacturing, Finance, IT
  • GPU kernels: DFG construction, pattern detection, partial order derivation, conformance checking

Changes

  • Updated showcase documentation with AccNet and ProcInt sections
  • Updated CI workflow to exclude CUDA tests on runners without GPU hardware

Fixes

  • Fixed 14 clippy warnings in ringkernel-accnet
  • Fixed benchmark API compatibility in ringkernel-accnet
  • Fixed code formatting issues across showcase applications

Run the Applications

# AccNet - Accounting Network Analytics
cargo run -p ringkernel-accnet --release

# ProcInt - Process Intelligence
cargo run -p ringkernel-procint --release

Full Changelog: v0.1.0...v0.1.1

Loading

RingKernel v0.1.0

03 Dec 16:12
@mivertowski mivertowski
v0.1.0
871e793

Choose a tag to compare

View all tags
RingKernel v0.1.0

RingKernel v0.1.0 - Initial Release

A GPU-native persistent actor model framework for Rust.

Highlights

  • Persistent GPU Kernels: GPU compute units as long-running actors that maintain state between invocations
  • Lock-free Message Queues: High-performance host↔GPU and kernel-to-kernel communication
  • Hybrid Logical Clocks (HLC): Causal ordering across distributed GPU operations
  • Multiple Backends: CPU, CUDA, WebGPU support
  • Zero-copy Serialization: rkyv-based message passing
  • Rust-to-GPU Transpilers: Write GPU kernels in Rust DSL, transpile to CUDA C or WGSL

Crates

Crate Description
ringkernel Main facade crate
ringkernel-core Core traits, types, HLC, K2K, PubSub
ringkernel-derive Proc macros (#[derive(RingMessage)], #[ring_kernel])
ringkernel-cpu CPU backend
ringkernel-cuda NVIDIA CUDA backend
ringkernel-wgpu WebGPU backend
ringkernel-cuda-codegen Rust-to-CUDA transpiler
ringkernel-wgpu-codegen Rust-to-WGSL transpiler
ringkernel-wavesim Wave simulation demo
ringkernel-txmon Transaction monitoring demo

Quick Start

[dependencies]
ringkernel = "0.1"
tokio = { version = "1", features = ["full"] }

For GPU backends:

ringkernel = { version = "0.1", features = ["cuda"] }
# or
ringkernel = { version = "0.1", features = ["wgpu"] }

Documentation

Performance

Benchmarked on NVIDIA RTX Ada:

  • CUDA Codegen: ~93B elem/sec (12,378x vs CPU)
  • Message queue throughput: ~75M ops/sec
  • HLC timestamp generation: <10ns per tick

What's Included

  • 14 workspace crates
  • 390+ tests
  • 20+ examples
  • Comprehensive documentation
  • Educational simulation modes (WaveSim)
  • Real-time fraud detection demo (TxMon)
Loading