The PRIME Research Parallel Real-time Immersive network Modeling Environment

Related Work

• PRIME SSF is the next generation C++ implementation of the Scalable Simulation Framework (SSF). It follows the design of DaSSF (or Dartmouth Scalable Simulation Framework) developed at Dartmouth College, and iSSF developed at the University of Illinois, Urbana-Champaign, primarily by the same author.

• PRIME SSF uses the Hoard memory allocator, optionally, when running on shared-memory multiprocessors. It also uses Doug Lea's memory allocator for managing the data segment.

• PRIME SSF can choose to use libSynk for synchronization between SSF instances running on distributed-memory machines for parallel and distributed discrete-event simulation. libSynk was written by Kalyan Perumalla when at Georgia Tech.

• PRIME SSF uses METIS, the graph partitioner by George Karypis at the University of Minnesota, for partitioning simulation models specified in Domain Modeling Language (DML).

• PRIME SSF uses two modules from the Boost library. One is the serialization library written by Robert Ramey, which is expected to be used by models that contain significant serialization code. The other is the random number generators that are included as part of the PRIME RNG library.

• PRIME SSF also uses the random number generators in the SPRNG (Scalable Parallel Random Number Generators) library. SPRNG is a project at the Florida State University and SPRNG library is distributed by NCSA.

• The multi-threading approach in PRIME SSF is inspired from the Cilk project at MIT.

• PRIME SSFNet is ported directly from the iSSFNet simulator, part of the MOSES project by Professor David Nicol's research group at the University of Illinois at Urbana-Champaign. The iSSFNet simulator was derived originally from the DaSSFNet simulator from Dartmouth College.

• The network emulation infrastructure in PRIME SSFNet is based on OpenVPN, a public-domain implementation of virtual private networks.