Distributed out-of-memory NMF on CPU/GPU architectures


We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10−6.

In The Journal of Supercomputing, 2023


NMF, out-of-memory, latent features , model selection , distributed processing , parallel programming, big data,heterogeneous computing , GPU , CUDA , NCCL , cupy


Boureima, I., Bhattarai, M., Eren, M. et al. Distributed out-of-memory NMF on CPU/GPU architectures. J Supercomput (2023). https://doi.org/10.1007/s11227-023-05587-4


  title={Distributed Out-of-Memory NMF on CPU/GPU Architectures},
  author={Boureima, Ismael and Bhattarai, Manish and Eren, Maksim and Skau, Erik and Romero, Philip and Eidenbenz, Stephan and Alexandrov, Boian},
  booktitle={The Journal of Supercomputing}, 
Maksim E. Eren
Maksim E. Eren

My research interests lie at the intersection of the machine learning and cybersecurity disciplines, with a concentration in tensor decomposition.