Interactive Supercomputing on 40,000 Cores for Machine Learning and Data Analysis

TL;DR

This paper presents the development of an interactive supercomputing system capable of launching thousands of machine learning and data analysis tasks within seconds on a 40,000-core supercomputer, enabling rapid experimentation.

Contribution

It introduces techniques for scaling interactive frameworks like TensorFlow and MATLAB/Octave to tens of thousands of cores with minimal latency.

Findings

32,000 TensorFlow processes launched in 4 seconds

262,000 Octave processes launched in 40 seconds

Enables rapid exploration of machine learning architectures

Abstract

Interactive massively parallel computations are critical for machine learning and data analysis. These computations are a staple of the MIT Lincoln Laboratory Supercomputing Center (LLSC) and has required the LLSC to develop unique interactive supercomputing capabilities. Scaling interactive machine learning frameworks, such as TensorFlow, and data analysis environments, such as MATLAB/Octave, to tens of thousands of cores presents many technical challenges - in particular, rapidly dispatching many tasks through a scheduler, such as Slurm, and starting many instances of applications with thousands of dependencies. Careful tuning of launches and prepositioning of applications overcome these challenges and allow the launching of thousands of tasks in seconds on a 40,000-core supercomputer. Specifically, this work demonstrates launching 32,000 TensorFlow processes in 4 seconds and launching 262,000 Octave processes in 40 seconds. These capabilities allow researchers to rapidly explore novel machine learning architecture and data analysis algorithms.