On this page... (hide)
Parametric computational experiments are becoming increasingly important in science and engineering as a means of exploring the behavior of complex systems. For example, an engineer may explore the behaviour of a wing by running a computational model of the airfoil multiple times while varying key parameters such as angle of attack, air speed, etc. The results of these multiple experiments yield a picture of how the wing behaves in different parts of parametric space. Over the past several years, we have developed a specialized parametric modeling system called Nimrod. Nimrod uses a simple declarative parametric modeling language to express a parametric experiment and provides machinery that automates the task of formulating, running, monitoring, and collating the results from the multiple individual experiments. Equally important, Nimrod incorporates a distributed scheduling component that can manage the scheduling of individual experiments to idle computers in a local area network. Together, these features mean that even complex parametric experiments can be defined and run with little programmer effort. In many cases it is possible to establish a new experiment in minutes.
|Nimrod/G||provides two services: Parameter sweeps and grid/cloud execution tools including scheduling across multiple compute resources. A commercial version of Nimrod, called EnFuzion, is available for clusters from|
|Nimrod/O||provides an optimisation framework for optimising a target output value of an application. Used with Nimrod/G, it can exploit parallelism in the search algorithm.|
|Nimrod/OI||provides an interactive interface for Nimrod/O. In some applications, it might require someone to decide which output is better. Those results are fed back into Nimrod/O to produce more suggestions.|
|Nimrod/E||provides experimental design techniques for analysing parameter effects on an application's output. Used with Nimrod/G allows the experiment to be scaled up on grid and cloud resources.|
|Nimrod/K||provides all the Nimrod tools in a workflow engine called. Nimrod/K adds all the parameter tools and grid/cloud services to Kepler while leveraging and enhancing all the existing grid tools already provided by adding dynamic parallelism in workflows.|
- Vary parameters
- Execute programs
- Copy data in and out
- Sequential and parallel dependencies
- Multiple parameter exploration tools, including sweeps, optimizations and experimental designs
- Computational economy drives scheduling
- Computation scheduled near data when appropriate
- Uses distributed high performance platforms
- Upper middleware broker for resources discovery
- Wide Community adoption
- Cloud computing interface
If you have used Nimrod for some research and would like to acknowledge it in your paper, we would welcome your use of the following text:
We wish to acknowledge Monash University for the use of their Nimrod software in this work. The Nimrod project has been funded by the Australian Research Council and a number of Australian Government agencies, and was initially developed by the Distributed Systems Technology CRC.
Nimrod Family references should cite: Abramson, D., Bethwaite, B., Enticott, C., Garic, S. and Peachey, T. “Parameter Exploration in Science and Engineering using Many-Task Computing”, Special issue of IEEE Transactions on Parallel and Distributed Systems on Many-Task Computing, June 2011, Volume: 22 Issue: 6, 960 – 973.
Nimrod/G references should cite: Abramson, D., Giddy, J. and Kotler, L. “High Performance Parametric Modeling with Nimrod/G: Killer Application for the Global Grid?”, International Parallel and Distributed Processing Symposium (IPDPS), pp 520- 528, Cancun, Mexico, May 2000
Nimrod/O references should cite: Abramson D, Lewis A, Peachey T, Fletcher, C., “An Automatic Design Optimization Tool and its Application to Computational Fluid Dynamics”, SuperComputing 2001, Denver, Nov 2001.
Nimrod/K references should cite: Abramson D, Enticott C, Altintas I., "Nimrod/K: Towards Massively Parallel Dynamic Grid Workflows", IEEE SuperComputing 2008, Austin, Texas November 2008