Siddharth Samsi

TL;DR: In this paper, the authors demonstrate the capabilities of a 40,000-core supercomputer to launch tens of thousands of TensorFlow and MATLAB/Octave processes in 40 seconds.

TL;DR: This paper surveys the current state of processors and accelerators that have been publicly announced with performance and power consumption numbers, and selects and benchmark two commercially available low size, weight, and power (SWaP) accelerators as these processors are the most interesting for embedded and mobile machine learning inference applications that are most applicable to the DoD and other SWaP constrained users.

TL;DR: In this paper, the authors show how to scale interactive machine learning frameworks, such as TensorFlow and MATLAB/Octave, to tens of thousands of cores by fine-tuning the launches and prepositioning of applications on a 40,000-core supercomputer.

TL;DR: This paper collects and summarizes the current accelerators that have been publicly announced with performance and power consumption numbers and highlights interesting trends regarding power consumption, numerical precision, and inference versus training.

TL;DR: The proposed Subgraph Isomorphism Graph Challenge draws upon prior challenges from machine learning, high performance computing, and visual analytics to create a graph challenge that is reflective of many real-world graph analytics processing systems.