EAGER: Kinetic Computing Sculpture: A functional parallel cluster of Raspberry Pi computers that inspire computational thinking

点击翻译按钮获取中文摘要

Without parallel computational thinking most modern scientific discoveries are impracticable (e.g. sequencing the human genome, confirmation of the Higgs boson). To recruit young minds at an early age (K-12) into the workforce and to encourage governments to continue investment, the parallel computing field must constantly educate non-computer scientists of all ages as to the importance of innovation that uncovers and exploits new capabilities. Unfortunately, educating laypersons as to the importance of parallel systems that grow exponentially large and complicated and change rapidly is wrought with challenges. Distribution of data and computation across a large parallel or distributed system is carried out by complex computer software. While the algorithms and systems used may have taken years to perfect, other than binary data packets and logs, and perhaps the hum of a system, there is little indication of the transmissions. Only a deeply informed individual is privy to the significance of the events taking place on a large parallel or distributed system.In this interdisciplinary project, the PIs are exploring radically different ways to improve our ability to educate non-computer scientists (e.g. students, K-12, funding decision makers, elderly, etc.) as to the importance and sophistication of parallel computation. This transformative approach combines the creation of a parallel or distributed system with mechanisms to visualize computation. With the combined expertise of computer scientists, educators, and artists, the PIs are designing and building a cluster of Raspberry Pi computers in a visually compelling, kinetic sculpture. Each computing device is attached to a compact servo mechanism fitted to an aluminum structure that will react to both computation and data movement of parallel programs running on the system. The PIs are designing displays that allow visitors to the sculpture to interactively select and view the effects of various parallel computation choices that computer and computational scientists make every day. The effectiveness of the proposed approach will be evaluated.