Computing Media and Languages for Space-Oriented Computation

Computing Media and Languages for Space-Oriented Computation

计算机科学;物理学

Fréderic ; Jean-Louis ; André

Others  :  http://drops.dagstuhl.de/opus/volltexte/2007/1025/pdf/06361.SWM.Paper.1025.pdf
PID  :  10312

学科分类：计算机科学（综合）

来源: CEUR

【 摘 要 】

Traditional models of computation have abstracted out physical locations in space (e.g. the Internet, superscalar processors, unit delay wires, uniform memory delay) and implementations predominantly perform computations in time (i.e. sequentially). Most of our common data structures are spatially agnostic (e.g. arrays). But: 1. As scaling continues (both as primitive elements shrink to the atomic scale, and the number of elements composed scales up), computations must be distributed in space and location in space impacts the performance and feasibility of the computation. 2. As we couple and embed computing in the physical world (e.g. smart building, reactive surfaces, programmable matter, distributed robotics), position and shape are primary, serving as both the input to computation and a key part of the desired result of the computation. 3. As we understand natural computing systems (e.g. cells, ant colonies, system’s biology) location and topology define the computation. Consequently, it is important to make space not an issue to abstract away, but a first-order effect that we optimize. The distinguishing feature of spatial computing then is that computation is performed distributed in space and position and distance metrics matter to the computation.[First Paragragh]