Timezone: »
We describe a new software framework for fast training of generalized linear models. The framework, named Snap Machine Learning (Snap ML), combines recent advances in machine learning systems and algorithms in a nested manner to reflect the hierarchical architecture of modern computing systems. We prove theoretically that such a hierarchical system can accelerate training in distributed environments where intra-node communication is cheaper than inter-node communication. Additionally, we provide a review of the implementation of Snap ML in terms of GPU acceleration, pipelining, communication patterns and software architecture, highlighting aspects that were critical for achieving high performance. We evaluate the performance of Snap ML in both single-node and multi-node environments, quantifying the benefit of the hierarchical scheme and the data streaming functionality, and comparing with other widely-used machine learning software frameworks. Finally, we present a logistic regression benchmark on the Criteo Terabyte Click Logs dataset and show that Snap ML achieves the same test loss an order of magnitude faster than any of the previously reported results, including those obtained using TensorFlow and scikit-learn.
Author Information
Celestine Dünner (IBM Research)
Thomas Parnell (IBM Research)
Dimitrios Sarigiannis (IBM Research)
Nikolas Ioannou (IBM Research)
Andreea Anghel (IBM Research)
Gummadi Ravi (IBM Systems)
Madhusudanan Kandasamy (IBM Systems)
Haris Pozidis (IBM Research)
More from the Same Authors
-
2020 Poster: SnapBoost: A Heterogeneous Boosting Machine »
Thomas Parnell · Andreea Anghel · Małgorzata Łazuka · Nikolas Ioannou · Sebastian Kurella · Peshal Agarwal · Nikolaos Papandreou · Haralampos Pozidis -
2019 Poster: SySCD: A System-Aware Parallel Coordinate Descent Algorithm »
Nikolas Ioannou · Celestine Mendler-Dünner · Thomas Parnell -
2019 Spotlight: SySCD: A System-Aware Parallel Coordinate Descent Algorithm »
Nikolas Ioannou · Celestine Mendler-Dünner · Thomas Parnell -
2017 Poster: Efficient Use of Limited-Memory Accelerators for Linear Learning on Heterogeneous Systems »
Celestine Dünner · Thomas Parnell · Martin Jaggi