This project will deploy an optically networked systems research infrastructure named BOLD that integrates high-performance, low-power, optical networking devices and programmable packet switches to enable transformative and inter-disciplinary research. It enables experimental exploration of different architectural design choices and it will potentially suggest new approaches to integrated hardware-software designs. At the hardware level, it enables research on novel optical network devices that can provide powerful communication capabilities, possibly tailored to the specific needs of big data applications, as well as experimentation with the device prototypes under real application traffic. BOLD enables a broad range of transformative big data-driven research. At the system software level, it enables research on storage, network, and application control software that are designed from the ground up to coordinate for optimal performance. At the application level, BOLD motivates research on how fundamental algorithms for big data applications should be designed to leverage the new network capabilities. BOLD has the unique potential to bridge the gap between nano-photonics researchers, networked systems researchers, and big data application researchers, creating inter-disciplinary research opportunities. Furthermore, because BOLD is designed to operate alongside Rice?s existing NSF-funded computing infrastructures BOLD can support big data experiments at a substantial scale.
Results from the inter-disciplinary research enabled by BOLD will lead to future big data processing system architectures that dramatically speed up a wide range of computational scientific discoveries. Because optical networking devices are unique in that they consume very little power, yet can support enormous data rates, BOLD will inspire a new class of high performance, high energy efficiency system architectures. Research enabled by BOLD could inform the design of future nation-wide networking infrastructures by showing how optical networks can be harnessed as shared ?cloud? resources. BOLD will serve as a platform for the training and education of numerous undergraduate and graduate students, including under-represented groups, in cutting edge big data-driven research.
