Hardware and Software Solutions for Energy-Efficient Computing in Scientific Programming

Energy consumption is one of the major issues in today’s computer science, and an increasing number scientific communities are interested evaluating tradeoff between time-to-solution energy-to-solution. Despite, last two decades, computing which revolved around centralized infrastructures, such as supercomputing data centers, wide adoption Internet Things (IoT) paradigm currently inverting this trend due to huge amount it generates, pushing power back places where generated—the so-called fog/edge computing. This shift towards a decentralized model requires equivalent change software engineering paradigms, development environments, hardware tools, languages, computation models for programming because local computational capabilities typically limited require careful evaluation consumption. paper aims present how these concepts can be actually implemented by presenting state art powerful, less power-hungry processors from side energy-aware tools techniques other one.