Modeling Of SEcurity and Systems (MOSES) iSSF Home Page http://www.crhc.uiuc.edu/~jasonliu/projects/issf/index.html

This project has been completed and therefore the project web page is no longer maintained actively. It is mirrored here for historical purposes. Check out www.primessf.net for new development in SSF.

Introduction

iSSF, formerly known as DaSSF, is a public-domain high-performance parallel and distributed simulator, designed to provide scalable simulation and emulation of large-scale complex systems. iSSF (pronounced as ice-if) is a portable C++ implementation of Scalable Simulation Framework, a public standard for process-oriented parallel simulation of large-scale communication networks.

Throughout the design and implementation of iSSF, we focus on speed and memory efficiency. We use conservative parallel simulation techniques that support simulation on both shared-memory multiprocessors and distributed-memory clusters of workstations. We use handcrafted threading mechanism to support light-weight process-oriented simulation paradigm. From the user's point of view, iSSF provides a simplistic and modular API that enables easy development of large-scale and complex models. iSSF is ultra fast and has been demonstrated capable of simulating extremely large network models. Models developed with iSSF are portable among a wide variety of hardware architectures, supporting a combination of shared-memory and distributed memory platforms.

iSSF is currently under active development to add support for HLA interoperability, real-time simulation, and human-interaction capabilities. The High Level Architecture (HLA) interface enables iSSF to interoperate with other existing simulators. Real-time simulation is to run the simulator (as a software program) to interact, in real time, with the physical world (through well-defined interfaces), so that, from an external standpoint, the simulator's function cannot be distinguished from the simulated system. In particular, real-time network simulation allows real traffic to go through the simulator and interact with simulated network entities. This will (i) promote studies of end-systems under controllable and repeatable network conditions, and (ii) help understand network behaviors under realistic application traffic patterns. We will also add user interface to simulation/emulation so that one can provide on-line control to the simulated system. Stay tuned for news that will be published at this web site.

iSSF is part of Modeling of Security Systems (MOSES) Project, at the Center for Reliable and High-Performance Computing, Coordinated Science Laboratory, University of Illinois, Urbana-Champaign.

News

• [09/25/2003] This web site is created.

Download

• Stay tuned for the first release of iSSF.
• Previous versions of DaSSF can be found at http://www.crhc.uiuc.edu/~jasonliu/projects/ssf/download.html.

People

• David M. Nicol, Professor, EECS, UIUC.
• Jason Liu, Postdoctoral Research Associate, CSL, UIUC.

Publications

2003

• Development of an internet backbone topology for large-scale network simulations, Michael Liljenstam, Jason Liu, and David M. Nicol. In Preceedings of the 2003 Winter Simulation Conference, New Orleans, LA, December 7-10, 2003.
• Simulation of large-scale networks using SSF, David M. Nicol, Jason Liu, Michael Liljenstam, and Guanhua Yan. In Preceedings of the 2003 Winter Simulation Conference, New Orleans, LA, December 7-10, 2003.
• Multiscale modeling and simulation of worm effects on the internet routing infrastructure, David M. Nicol, Michael Liljenstam, and Jason Liu. In Proceedings of the 13th International Conference on Modelling Techniques and Tools for Computer Performance Evaluation (Performance TOOLS 2003), Urbana, IL, September 2-5, 2003.
• An implementation of the SSF Scalable Simulation Framework on the Cray MTA, Robert R. Henry, Simon H. Kahan, Jason Liu, and David M. Nicol, In Proceedings the 17th Workshop on Parallel and Distributed Simulation (PADS'03), San Diego, CA, June 10-13, 2003, pp. 77-85.
• Improvements in conservative parallel simulation of large-scale models, Xiaowen (Jason) Liu, Ph.D. Thesis, Dartmouth Collge, Hanover, NH, February 2003.

Selected old publications

• Composite synchronization in parallel discrete-event simulation, David M. Nicol and Jason Liu, IEEE Transactions on Parallel and Distributed Systems, 13(5), May 2002, pp. 433-446.
• Lookahead revisited in wireless network simulations, Jason Liu and David M. Nicol, In Proceedings of the 16th Workshop on Parallel and Distributed Simulation (PADS'02), Washington, D.C., May 12-15, 2002, pp. 79-88.
• Lock-free scheduling of logical processes in parallel simulation, Jason Liu, David M. Nicol, and King Tan, In Proceedings of the 15th Workshop on Parallel and Distributed Simulation (PADS'01), Lake Arrowhead, CA, May 15-18, 2001, pp. 22-31.
• Learning not to share, David M. Nicol and Jason Liu, In Proceedings of the 15th Workshop on Parallel and Distributed Simulation (PADS'01), Lake Arrowhead, CA, May 15-18, 2001, pp. 46-55.
• Safe timestamps and large-scale modeling, David M. Nicol, Jason Liu, and James Cowie, In Proceedings of the 14th Workshop on Parallel and Distributed Simulation (PADS'00), Bologna, Italy, May 28-31, 2000, pp. 71-78.
• Towards realistic million-node Internet simulations, James Cowie, Hongbo Liu, Jason Liu, David Nicol and Andy Ogielski, In Proceedings of the 1999 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA'99), Las Vegas, NV, June 28-July 1, 1999.
• Performance prediction of a parallel simulator, Jason Liu, David M. Nicol, Brian J. Premore, and Anna L. Poplawski, In Proceedings of the 13th Workshop on Parallel and Distributed Simulation (PADS'99), Atlanta, GA, May 1-4, 1999, pp. 156-164.

Links

• Modeling of Security Systems (MOSES) Project.
• Center for Reliable and High-Performance Computing, at Coordinated Science Laboratory, University of Illinois, Urbana-Champaign.
• Dartmouth Scalable Simulation Framework (DaSSF).
• SSF Research Network.