FloatSOM: GPU-Accelerated, Distributed, Topology-Flexible Self-Organizing Maps

TL;DR

FloatSOM is a scalable GPU-accelerated framework for self-organizing maps that supports large datasets, distributed execution, and flexible topologies, achieving state-of-the-art accuracy and efficiency.

Contribution

It introduces FloatSOM, enabling multi-GPU, out-of-memory streaming, and novel topologies for large-scale SOM analysis with improved accuracy.

Findings

Lower quantization error than current SOM baselines.

Trains a 1024-node SOM on 1 billion samples in 6.16 minutes.

Supports large-scale, distributed, topology-flexible SOM training.

Abstract

GPU-accelerated Self-Organizing Map (SOM) implementations are among the most competitive options for large-scale SOM analysis, but growing dataset sizes increasingly challenge their practical use because workloads no longer fit cleanly within device-memory limits. We introduce FloatSOM, a SOM framework for scalable training and deployment that supports multi-GPU execution, out-of-memory disk-backed streaming, and novel topologies beyond regular lattices. We evaluate FloatSOM on 14 synthetic and real benchmark datasets together with controlled speed scaling benchmarks, and show that these improved topologies, combined with topology-aware hyperparameter fine-tuning, yield lower quantization error than current state-of-the-art SOM baselines. FloatSOM also sustains this performance at large scale with high-throughput distributed execution; in the largest benchmark, it trains a 1024-node SOM network on 1,000,000,000 samples with 50 features in 6.16 minutes on 8 GPUs across two separate high-performance-computing nodes.