README.md

metadata

title: FlexChunk SpMV Demo
emoji: ⚡
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.39.0
app_file: app.py
pinned: false

FlexChunk: Enabling Massive Out-of-Core SpMV

This interactive demo showcases FlexChunk, an algorithm for performing Sparse Matrix-Vector Multiplication (SpMV) on matrices too large to fit in RAM.

Key Benefit: Process matrices up to 300M×300M using only ~1.7 GB RAM by dividing them into manageable horizontal chunks.

Interactive Demo

The application above provides two modes:

• Standard Mode: For matrices up to 200K×200K with optional SciPy comparison
• Advanced Mode: For matrices up to 300M×300M using FlexChunk only

You can adjust parameters including:

• Matrix size
• Density
• Number of chunks
• Matrix type (standard or challenging with extreme values)

Performance Highlights

FlexChunk demonstrates near-linear scaling in both time and memory usage:

Matrix Size	Non-zero Elements	Total Time	Peak RAM Usage
1.0M × 1.0M	1.2M	1.07 s	17.00 MB
10.0M × 10.0M	12.0M	10.21 s	170.00 MB
50.0M × 50.0M	62.5M	55.27 s	850.00 MB
100.0M × 100.0M	120.0M	1 min 47.1 s	1.70 GB

Our algorithm scales linearly to even larger matrices (up to 300M×300M) with proportional increases in processing time and memory.

How It Works

FlexChunk operates in three main stages:

1. Matrix Division: Splits the matrix into horizontal chunks and saves to disk
2. Sequential Processing: Loads one chunk at a time to minimize memory usage
3. Result Accumulation: Combines partial results into the final vector

This approach makes it possible to multiply vectors with matrices that would otherwise exceed available RAM.

Links

• Source Code: GitHub Repository
• Full Article: FlexChunk: Enabling 100M×100M Out-of-Core SpMV