High speed bulk data transfer using the SSH protocol

  • Authors:
  • Chris Rapier;Benjamin Bennett

  • Affiliations:
  • Pittsburgh Supercomputing Center, Pittsburgh, PA;Pittsburgh Supercomputing Center, Pittsburgh, PA

  • Venue:
  • Proceedings of the 15th ACM Mardi Gras conference: From lightweight mash-ups to lambda grids: Understanding the spectrum of distributed computing requirements, applications, tools, infrastructures, interoperability, and the incremental adoption of key capabilities
  • Year:
  • 2008

Quantified Score

Hi-index 0.01

Visualization

Abstract

SSH is a highly successful multipurpose protocol used for both interactive shells and transport layer mechanisms. However, a design choice in most implementations of SSH reduces its functionality as bulk data transport tool in high performance network environments. This paper will discuss the nature of this limitation, the functional barriers it imposes, a method by which it can be remedied, and introduces a high performance implementation based on the industry standard, OpenSSH. Additionally, the authors will introduce a method by which performance on multi-core systems is improved through the use of cipher multi-threading.