Isoenergy maps show relationships between pairs of the system and application parameters which must hold to keep energy used in the parallel computation constant. For example, speed of communication (i.e. 1/C) should increase with incerasing speed of computation (1/A). However, this relationship has peculiar knee-like shape. When energy wasted in computation is small (becasue computations are fast), then the overal energy consumption becomes insensitive to the speed of computation, and then the communication time must be also reduced (by incresaing speed 1/C). The same applies to the speed of communication. The effective optimization in one corner (e.g. computation speed 1/A) exposes deficiencies in the other corner (1/C). Thus both speeds must be optimized in unison. In the complex system of relationships determining overall performance of the parallel computation, isolines of the kind present in the isoenergy maps help build understanding of the correlations between key parameters, their sensitivites, and limitations of such correlations.

More detailed explanation of the idea of isoenergy lines can be found in:
Maciej Drozdowski, Jędrzej M. Marszałkowski, Jakub Marszałkowski Energy trade-offs analysis using equal-energy maps, Future Generation Computer Systems, 2013, accepted, http://dx.doi.org/10.1016/j.future.2013.07.004

The applet for isoenergy modeling was first implemented in Java by Mateusz Dukowski, Michał Gałka, Adam Ulatowski, Michał Wieliński, as part of their Bachelor in Science thesis in computer science in 2011. Then the applet was ported to JavaScript by Jakub Białek in 2019 as a partial fulfilment of his MSci thesis.