Tapas K. Som

A Formal Development of Event Graphs as an Aid to Structured and Efficient Simulation Programs

Schruben introduced event graphs as a graphical technique for visualizing event-oriented system structures. In this paper, a formal framework for event-oriented simulation is introduced and event graphs are examined using this framework. Comprehensive rules are presented for identifying situations where event execution order priorities may be necessary. The concept of expanded event graphs and super events are introduced. A procedure for generating expanded event graphs which show hierarchy and possible grouping of events as one single unit without violating predefined execution order priorities is described. Expanded event graphs generated by this procedure can be used as bases for structured and efficient simulation programs. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

A probabilistic event scheduling policy for optimistic parallel discrete event simulation

A probabilistic scheduling policy for selecting the next event to be processed by a processor during optimistic parallel discrete event simulation is presented. This policy exploits the knowledge about statistical distributions of the difference between the timestamps of successively committed events of a logical process. Simulation results are presented to demonstrate that under appropriate conditions, probabilistic scheduling generates substantially fewer rollbacks compared to the lowest timestamp first scheduling policy.

Model structure and load balancing in optimistic parallel discrete event simulation

The concept of strong groups is introduced to describe the structure of simulation models. It is shown that logical processes within strong groups process at approximately the same rate and that different strong groups can progress at different rates. An algorithm based on the rates of the strong groups is presented to balance the load among the physical processors and for flow control.

On Analyzing Events To Estimate The Possible Speedup Of Parallel Discrete Event Simulation

We describe in this paper a new event based approach for estimating the possible speed up in parallel discrete event simulation which is similar to Berry and Jeffersons method. This method calculates a critical path through a graph developed from a trace of a simulation and from constraints on the order of processing of event instances that ensure correctness of the simulation. Different ways of classifying event instances to determine required constraints are presented. A software system that uses this approach is described and examples are presented.

Frequency domain metamodelling of a feedback queue

Schruben and Cogliano [1987] introduced Frequency Domain Experiments as a tool for metamodel identification. To design a frequency domain experiment, the experimenter must choose appropriate values for certain experimental variables such as oscillation frequency, window size, and range of oscillation. In this paper, we demonstrate that these experimental variables affect the outcome of frequency domain experiments and the magnitude of these effects are model dependent.

A New Process to Processor Assignment Criterion for Reducing Rollbacks in Optimistic Simulation

A new criterion for assigning logical processes to processors during optimistic parallel discrete event simulation runs is presented. Instead of assigning closely coupled logical processes to the same processor, the logical processes experiencing rollback caused by a common source (another logical process) are assigned to the same processor. This approach significantly reduces the number of rollbacks. An algorithm to generate the assignments based on this new criterion is also presented.

Alternatate methods for generating and analyzing the output series of frequency domain experiments

In this paper we compare two different methods of spectral estimates for Frequency Domain Experiments and present some performance data for execution on a SUN 3/50 workstation. We also suggest an alternative method for generating the output series by replication and demonstrate that at least for some cases this is more efficient than using a single run of longer length.