IMSuite: IIT Madras Benchmark Suite for Simulating
IMSuite includes a problem specific input generator for
twelve classical distributed algorithms
The presence of such an input generator allows creation of
input sets necessary for each benchmark kernel.
These input generators write the desired input to a file
and these files then act as the input to the kernels.
Each input generator permits a set of options which can be
passed during the runtime.
In case an option is not used, the default value is used.
Some of the options available with various generators are:
- -in - For specifying the input file name.
- -sz - For specifying the size of input.
- -rn - For generating a random graph.
- -cplt - For generating a complete graph.
- -spmax - For generating a Sparse maximum graph.
- -spmin - For generating a Sparse minimum graph.
- -star - For generating a tree with star configuration.
- -chain - For generating a tree with chain configuration.
- -seed - For specifying a new seed for input generation.
Options specific to an input generator are stated as comments inside each input generator code.
The Input Generators can be downloaded here
Details regarding generation of an input is present here
The IMSuite distribution currently consists of input sets with seven different
sizes, ranging from 8 to 512 in multiples of two.
These sizes refer to the number of nodes in a network.
Larger input sets can also be generated.
The default seed is set to 101.
To download individual input sets (generated with default seed) click on the appropriate Input Size:-
The collective input sets are available here
last modified 11:21, 25 April 2014