IMSuite: IIT Madras Benchmark Suite for Simulating
Distributed Algorithms

IMSuite Kernel Categorization

IMSuite include implementation for twelve classical distributed algorithms. These implementations can be broadly subdivided under following two heads: (Version date: 21 December 2014)

• Timing -   These kernels help in measuring the time taken by a benchmark kernel to execute.
Select a version HJ: hj-1.3.1 X10: x10-2.3.0 X10: x10-2.5.0 OpenMP: OpenMP 4
• Characterization - These kernels help in analyzing the characteristics pertinent to the distributed and parallel systems.
Select a version HJ: hj-1.3.1 X10: x10-2.3.0 X10: x10-2.5.0

Based on the different form of parallelization, the kernels present in the aforementioned heads are further subdivided under two categories: Data parallel kernels and Recursive task parallel kernels.

Full IMSuite (including Inputs) -- Download

Details regarding IMSuite kernel benchmarks can be studied here.
Details regarding compilation and execution of IMSuite kernels are available here.

---

Change Log