Optics has the potential to solve some of the most exciting problems in information systems. It promises crosstalk-free interconnects with essentially unlimited bandwidth, long-distance data transmission without skew and without power- and time-consuming regeneration, miniaturization, parallelism, and efficient implementation of important algorithms such as Fourier transforms. Numerous information processing systems and concepts in space and time have been studied during the past decades. Yet, optics in computing and information processing in space and time has so far failed to move out of the lab. The current optical technology is costly, bulky, fragile in their alignment, and difficult to integrate with electronic systems, both in terms of the fabrication process and in terms of delivery and retrieval of massive volumes of data the optical elements can process. Our most recent work emphasizes the construction of optical subsystems directly on-chip, with the same lithographic tools as the surrounding electronics. This has been made possible by the advances in these tools, which can now create features significantly smaller than the optical wavelength, which is predicted to reach as fine as 11 nm by 2020. Arranged in a regular pattern, subwavelength features they act as a metamaterial whose optical properties are controlled by the density and geometry of the pattern and its constituents [1–5]. Moreover, composite metal-dielectric-semiconductor gain geometries [6] are used to create a new type of nanolasers for information systems integration.