SAGA

Project Information

Discipline
Computer Science (401) 
Orientation
Research 
Abstract

The Simple API for Grid Applications (SAGA) is an OGF standard (http://www.ogf.org), and defines a high level, application oriented API for developing first principle distributed applications. Our SAGA implementation (in C++ and Python, see http://saga.cct.lsu.edu/) is able to interface to a variety of middleware backends. We also develop application frameworks based on SAGA, such as Master-Worker, MapReduce, AllPairs, BigJobs, etc.\n\n\n\nFor all those components, we intent to use futuregrid and the different software environments available on FG for extensive portability and interoperability testing, but also for scale-up and scale-out experiments. The proposed activities will allow to harden the SAGA components described above.

Intellectual Merit

The Simple API for Grid Applications (SAGA) is an OGF standard (http://www.ogf.org), and defines a high level, application oriented API for developing first principle distributed applications. Our SAGA implementation (in C++ and Python, see http://saga.cct.lsu.edu/) is able to interface to a variety of middleware backends. We also develop application frameworks based on SAGA, such as Master-Worker, MapReduce, AllPairs, BigJobs, etc.\n\n\n\nFor all those components, we intent to use futuregrid and the different software environments available on FG for extensive portability and interoperability testing, but also for scale-up and scale-out experiments. The proposed activities will allow to harden the SAGA components described above.

Broader Impacts

The Simple API for Grid Applications (SAGA) is an OGF standard (http://www.ogf.org), and defines a high level, application oriented API for developing first principle distributed applications. Our SAGA implementation (in C++ and Python, see http://saga.cct.lsu.edu/) is able to interface to a variety of middleware backends. We also develop application frameworks based on SAGA, such as Master-Worker, MapReduce, AllPairs, BigJobs, etc.\n\n\n\nFor all those components, we intent to use futuregrid and the different software environments available on FG for extensive portability and interoperability testing, but also for scale-up and scale-out experiments. The proposed activities will allow to harden the SAGA components described above.

Project Contact

Project Lead
Shantenu Jha (shantenu) 
Project Manager
Yaakoub El Khamra (yye00) 
Project Members
Ole Weidner, Hartmut Kaiser, Pradeep Kumar Mantha, Andre Luckow, Sharath Maddineni, Pradeep Kumar Mantha, Andre Merzky, Melissa Romanus, Ashley Zebrowski, Mark Santcroos, Blaise Bourdin, Frank Löffler, Ole Weidner, Vishal Shah, Anjanibhargavi Ragothaman, Jeffrey Rabinowitz, Dinesh Ganapathi, Sanket Wagle, Matteo Turilli, Christian Straube, Antons Treikalis, Michael Wilde, Antons Treikalis, Nikhil Shenoy, Daniel S. Katz  

Resource Requirements

Hardware Systems
  • alamo (Dell optiplex at TACC)
  • foxtrot (IBM iDataPlex at UF)
  • hotel (IBM iDataPlex at U Chicago)
  • india (IBM iDataPlex at IU)
  • sierra (IBM iDataPlex at SDSC)
  • xray (Cray XM5 at IU)
  • bravo (large memory machine at IU)
  • delta (GPU Cloud)
  • Network Impairment Device
  • Not sure
 
Use of FutureGrid

development, porting and testing of SAGA (see web page)\n\n - scale-out and scale-up experiments of SAGA based applications and frameworks\n\n - interoperability testing for distributed SAGA based applications (see also http://forge.gridforum.org/sf/projects/gin)

Scale of Use

In general we have very low scale requirements, but would like to be able to test scale up and scale-out now and then, for very short periods of time (hours to a few days).

Project Timeline

Submitted
10/15/2010 - 02:13