Auto-pipe: Streaming applications on architecturally diverse systems

Roger D. Chamberlain, Mark A. Franklin, Eric J. Tyson, James H. Buckley, Jeremy Buhler, Greg Galloway, Saurabh Gayen, Michael Hall, E. F.Berkley Shands, Naveen Singla

Research output: Contribution to specialist publicationArticle

23 Scopus citations

Abstract

Auto-Pipe, an application development environment for streaming applications executing on architecturally diverse computing platforms, supports the flexible mapping and automatic delivery of application components between computational resources. Auto-Pipe's focus is on enabling designers to develop high-performance applications that run correctly despite the above limitations. An important component of this development environment is the emphasis placed on performance assessment and evaluation. The major purpose for deploying applications on diverse systems is to exploit the achievable performance gains. In addition to the computational finance application, Auto-Pipe has been used to implement applications ranging from cryptography to astrophysics. Auto-Pipe currently supports applications deployed on chip multiprocessors and FPGAs. In addition, a block library and incorporation of analog computation is under development.

Original languageEnglish
Pages42-49
Number of pages8
Volume43
No3
Specialist publicationComputer
DOIs
StatePublished - Mar 2010

Keywords

  • Auto-Pipe
  • Computer systems organization
  • Multicore processors
  • Processor architectures
  • Software engineering

Fingerprint

Dive into the research topics of 'Auto-pipe: Streaming applications on architecturally diverse systems'. Together they form a unique fingerprint.

Cite this