License: Creative Commons Attribution 3.0 Unported license (CC BY 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.OPODIS.2016.1
URN: urn:nbn:de:0030-drops-70705
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2017/7070/
Go to the corresponding LIPIcs Volume Portal

Shavit, Nir

High Throughput Connectomics (Keynote Abstract)

pdf-format:
LIPIcs-OPODIS-2016-1.pdf (0.2 MB)

Abstract

Connectomics is an emerging field of neurobiology that uses cutting edge machine learning and image processing to extract brain connectivity graphs from electron microscopy images. It has long been assumed that the processing of connectomics data will require mass storage and farms of CPUs and GPUs and will take months if not years. This talk will discuss the feasibility of designing a high-throughput connectomics-on-demand system that runs on a multicore machine with less than 100 cores and extracts connectomes at the terabyte per hour pace of modern electron microscopes. Building this system required solving algorithmic and performance engineering issues related to scaling machine learning on multicore architectures, and may have important lessons for other problem spaces in the natural sciences, where until now large distributed server or GPU farms seemed to be the only way to go.

BibTeX - Entry

@InProceedings{shavit:LIPIcs:2017:7070,
  author =	{Nir Shavit},
  title =	{{High Throughput Connectomics (Keynote Abstract)}},
  booktitle =	{20th International Conference on Principles of Distributed Systems (OPODIS 2016)},
  pages =	{1:1--1:1},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-031-6},
  ISSN =	{1868-8969},
  year =	{2017},
  volume =	{70},
  editor =	{Panagiota Fatourou and Ernesto Jim{\'e}nez and Fernando Pedone},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum fuer Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{http://drops.dagstuhl.de/opus/volltexte/2017/7070},
  URN =		{urn:nbn:de:0030-drops-70705},
  doi =		{10.4230/LIPIcs.OPODIS.2016.1},
  annote =	{Keywords: Machine learning, multicore architectures}
}

Keywords: Machine learning, multicore architectures
Collection: 20th International Conference on Principles of Distributed Systems (OPODIS 2016)
Issue Date: 2017
Date of publication: 06.04.2017

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI