HeAT -- a Distributed and GPU-accelerated Tensor Framework for Data Analytics
This addresses the bottleneck of single-node resource limitations for data scientists and researchers, enabling easier and faster large-scale data analysis, though it is incremental as it builds on existing technologies like PyTorch and MPI.
The authors tackled the problem of scaling data analysis and machine learning to distributed high-performance computing systems by introducing HeAT, a distributed and GPU-accelerated tensor framework with a NumPy-like API, achieving speedups of up to two orders of magnitude compared to similar frameworks.
To cope with the rapid growth in available data, the efficiency of data analysis and machine learning libraries has recently received increased attention. Although great advancements have been made in traditional array-based computations, most are limited by the resources available on a single computation node. Consequently, novel approaches must be made to exploit distributed resources, e.g. distributed memory architectures. To this end, we introduce HeAT, an array-based numerical programming framework for large-scale parallel processing with an easy-to-use NumPy-like API. HeAT utilizes PyTorch as a node-local eager execution engine and distributes the workload on arbitrarily large high-performance computing systems via MPI. It provides both low-level array computations, as well as assorted higher-level algorithms. With HeAT, it is possible for a NumPy user to take full advantage of their available resources, significantly lowering the barrier to distributed data analysis. When compared to similar frameworks, HeAT achieves speedups of up to two orders of magnitude.