Concentrator v2.0: GPU Accelerated Hashcat Rule Processor & Generator

New in v2.0 - Enhanced Features

🚀 Enhanced GPU Acceleration

GPU Extraction Mode
Now validates extracted rules on GPU

Enhanced Validation Kernel
validate_and_format_rules() with better output

Batch Processing
Mass parallel validation for all modes

💾 System & UI Enhancements

RAM Monitoring
Real-time memory usage tracking

Swap Space Detection
Automatic fallback warnings

Enhanced Interactive Mode
Better user experience

Memory Safety
Warnings for intensive operations

Performance Improvements

Extraction Mode: Now with GPU validation
Combinatorial Mode: 5-20x faster with GPU
Markov Mode: Optional GPU post-validation
Validation: 10-100x faster with GPU

Memory Management: Smart RAM monitoring
User Experience: Enhanced interactive mode
Output Quality: Better formatted rules
Error Handling: Improved fallback system

GPU Acceleration: What's Actually Accelerated

✅ GPU Accelerated

Combinatorial Generation
Full GPU rule creation with validation

Batch Rule Validation
Mass parallel syntax checking

Enhanced Extraction Validation
GPU validation for extracted rules

OpenCL Kernels
Custom GPU code for rule processing

❌ CPU Only

Markov Generation
Statistical rule creation

File Analysis
Rule extraction and counting

Model Building
Markov probability calculations

Interactive Mode Logic
User interface processing

🔄 Hybrid Processing

Markov Validation
GPU validation after CPU generation

Extraction Mode
CPU analysis + GPU validation

Interactive Mode
CPU UI with GPU backend

Fallback System
GPU → CPU on error

Performance Reality Check

Combinatorial Mode: 5-20x faster with GPU
Markov Mode: No GPU acceleration for generation
Validation: 10-100x faster with GPU
Extraction: CPU analysis + GPU validation

Best for GPU: Mass rule generation & validation
Best for CPU: Statistical analysis & file I/O
Hybrid Benefit: Validation across all modes
Fallback: Seamless CPU operation

Hybrid Architecture & Purpose

Concentrator v2.0 uses a smart hybrid approach: GPU for parallelizable tasks, CPU for sequential algorithms. This ensures optimal performance for each processing type.

GPU Tasks: Combinatorial generation, batch validation, extraction validation
CPU Tasks: Markov generation, file analysis, model building, interactive UI
Interactive Mode: Guided user interface with real-time validation
Smart Fallback: Automatic CPU operation if GPU unavailable

Technical Architecture

• OpenCL GPU Processing

• Multi-core CPU Processing

• Interactive CLI Interface

• Markov Models (CPU)

• Combinatorial Math (GPU)

• Hybrid Validation

• RAM Monitoring System

• External Cleanup Integration

Three Processing Modes - GPU/CPU Breakdown

Extraction Mode (-e)

• CPU: File analysis & sorting

• CPU: Markov model building

• GPU: Enhanced rule validation

• Multi-core optimized

• NEW: GPU validation for extracted rules

concentrator.py rules/ -e -t 10000

Primary: CPU | Validation: GPU Enhanced

Combinatorial Mode (-g)

• GPU: Mass rule generation

• GPU: Built-in validation

• CPU: Operator selection

• 5-20x faster with GPU

• GPU: Full combinatorial processing

concentrator.py rules/ -g -n 100000 -l 1 3

Primary: GPU | Fallback: CPU

Markov Mode (-gm)

• CPU: Statistical generation

• CPU: Probability calculations

• GPU Optional: Post-validation

• Context-aware sequences

• CPU: Sequential model traversal

concentrator.py rules/ -gm -gt 10000 -ml 1 5

Primary: CPU | Validation: GPU Optional

GPU OpenCL Acceleration Engine

OpenCL Kernel Features

Combinatorial Generation: Full GPU rule creation

Batch Validation: Thousands of rules in parallel

Enhanced Extraction: validate_and_format_rules() kernel

Memory Optimization: 16-byte aligned buffers

Platform Support

NVIDIA: CUDA through OpenCL

AMD: ROCm OpenCL support

Intel: Integrated GPU acceleration

CPU Fallback: Multi-core OpenCL

Auto-detection: Smart device selection

CPU-Only Components

Markov Generation:
Sequential probability-based traversal
Not suitable for GPU parallelization

File Analysis:
I/O-bound operations
Better suited for multi-core CPU

Model Building:
Complex statistical calculations
CPU-optimized algorithms

Interactive Mode:
User interface logic
Real-time user interaction

GPU Kernel Architecture

validate_rules_batch(): Parallel syntax checking

validate_and_format_rules(): Enhanced validation with formatting

generate_combinatorial_rules(): Mass rule generation

Lookup Tables: Operator requirements in constant memory

Thread Distribution: Each GPU thread processes multiple rules

Real GPU Usage Example

🎯 OpenCL initialized on: NVIDIA GeForce RTX 4090

🚀 GPU Acceleration: ENABLED

🎯 GPU generating 1,000,000 combinatorial rules...

✅ GPU generated 650,432 valid rules

🎯 GPU validating extracted rules...

✅ GPU validated 45,678 formatted rules

[Markov generation running on CPU...]

Smart RAM Monitoring & Memory Management

Memory Monitoring Features

Real-time RAM Tracking: Live memory usage monitoring

Swap Space Detection: Automatic swap usage detection

Warning System: Alerts for high memory usage

User Confirmation: Prompts for intensive operations

Fallback Warnings: Critical alerts for no swap space

Memory Safety

Check Before Intensive Ops: Validates available memory

In-Memory Mode: Optional RAM-only processing

Temp File Management: Disk-based fallback

Process Isolation: Multi-core memory optimization

Memory Status Output

📊 Memory Status: RAM 65.2% (12.4/19.1 GB)

💾 Swap available: 4.0 GB

⚠️ WARNING: High RAM usage detected (85.7%)

🚫 CRITICAL: No swap space available

Continue with memory-intensive operation? (y/N):

Enhanced Interactive Mode

Interactive Features

Guided Setup: Step-by-step configuration

File Selection: Interactive file/folder browser

Mode Selection: Detailed mode descriptions

Parameter Validation: Real-time input checking

Summary Display: Processing overview before execution

User Experience

Visual Feedback: Emoji-based status indicators

Progress Tracking: Real-time operation status

Error Handling: Clear error messages with solutions

Confirmation Prompts: Safe operation confirmation

Interactive Session Example

🎯 Active Mode: EXTRACTION

✅ File analysis complete: rules/best64.rule

📊 Total unique rules loaded into memory: 45,321

🎯 GPU validating extracted rules...

✅ GPU validated 10,000 formatted rules

💾 File 'concentrator_extracted.txt' saved

🔧 Running cleanup: ./cleanup-rules.bin 2

Hybrid Processing Pipeline

Phase 0: Interactive Configuration

User Input & Validation

• File path collection

• Mode selection with descriptions

• GPU/CPU preference settings

• Parameter validation

• RAM usage warnings

• Processing confirmation

Phase 1: Hybrid System Setup

GPU OpenCL + CPU Multiprocessing

# GPU Initialization
platforms = cl.get_platforms()
devices = platforms[0].get_devices(cl.device_type.GPU)
# CPU Initialization
pool = multiprocessing.Pool(processes=num_cores)
# RAM Monitoring
memory_info = check_ram_usage()

• GPU device detection

• OpenCL kernel compilation

• CPU core allocation

• RAM monitoring setup

• Fallback preparation

• Memory safety checks

Phase 2: CPU File Analysis

Multi-Core File Processing (CPU)

• Operator frequency counting

• Full rule occurrence tracking

• Markov model data collection

• I/O optimized operations

• Multi-core distribution

• Memory usage monitoring

Phase 3: Smart Mode Execution

Extraction Mode (Hybrid)

• CPU: Sort by frequency/Markov weight
• CPU: Top-N rule selection
• GPU: Enhanced rule validation
• Output: Properly formatted rules

Combinatorial Mode (GPU)

• GPU: Mass rule generation
• GPU: Built-in validation
• CPU: Operator selection logic
• GPU: Parallel processing
• Output: Validated rule sets

Markov Mode (Hybrid)

• CPU: Statistical generation
• CPU: Probability calculations
• CPU: Sequential traversal
• GPU: Optional post-validation
• Output: Statistically probable rules

Performance Optimizations

GPU Combinatorial Generation

Massively parallel rule creation and validation on GPU for combinatorial mode

Enhanced GPU Extraction

RAM Monitoring System

Real-time memory tracking with warnings for intensive operations

Interactive Mode

Guided user interface with visual feedback and validation