The edge computing domain is notably diverse in its composition. Capabilities including sensing, actuation, and mobility are often constrained by limitations such as those in energy, storage, and hardware-specific implementations. Leveraging the diversity of cloud-edge systems when building applications pose fundamental challenges that constrain expressivity, portability, and reconfigurability. Given this complexity, system tooling is necessary to aid application development and deployment in order to provide a stable runtime foundation by which environment dynamics can be accounted for.

This dissertation explores the notion of incorporating self-awareness into autonomous systems spanning the cloud and edge, such that they might observe the constantly changing state of their applications and resources to implicitly adapt for optimized behavior. Key thrusts include (1) DDFlow, a macroprogramming abstraction that organizes distributed applications into a visual dataflow representation, (2) Portkey, an adaptive key-value store that reconfigures data placement based on access patterns, and (3) EdgeRM, a distributed resource manager that unifies a heterogeneous computing cluster spanning high-performance servers and low-power sensors, which leverages a WebAssembly execution model and system interface extension for unified sensing.

Ultimately, this research seeks to lift IoT and embedded devices into the general-purpose computing domain, thereby enabling a carry-over of technical contributions pioneered by the distributed systems community. As systems researchers continue to introduce new paradigms in how to design, deploy, and support applications spanning cloud-edge, our community can help manage the difficulty in harnessing the capabilities of these networks.

When faced with starvation, animals become more aware of food-related stimuli, at the expense of other behaviors such as courtship. Underlying this heightened stimulus perception is the altered activity of the sensory systems, particularly the olfactory system. To study this, the model organism Drosophila was used due to its simple anatomical olfactory and neural systems, and its genetic tractability. We used a high-throughput single-fly assay to measure the effect of starvation duration on food search behavior. We found mild and severe starvation conditions recruit differential neural substrates for odor search behavior. A signaling pathway for short Neuropeptide F (sNPF) in the mushroom body, particularly in the γ lobe, is shown to be required for this severe starvation-induced food search behavior, where sNPF receptors are localized on

the γ1 and γ2 MBONs. Lastly, the increased excitability of the γ1 and γ2 mushroom body output neurons (MBONs) was able to induce severe starvation behavior in mildly starved flies reinforcing the state-dependent neural plasticity of Drosophila. These results suggest a sNPF-signaling pathway that modulates the activity of the mushroom body in order to increase the food search behavior of Drosophila as starvation progresses.

Blue and violet laser diodes (LDs) made from the (Al,In)GaN material system were first demonstrated in 1995 and have since been commercialized for applications in data torage and display technology. As the material and device technology continues to mature, these laser diodes are being investigated for use in solid state lighting and wireless communications, most recently being utilized for high-end automobile headlights. Competitiveness in these markets will require new device designs of GaN-based LDs to improve the efficiency and optical output power.

A key structural element of a laser diode is the set of cladding layers around a waveguiding core which, together, confine the optical mode. This confinement is dependent on the refractive index contrast between the cladding and core. However, it is difficult to grow lattice-mismatched AlGaN and InGaN layers with high enough composition and thickness to provide the required index contrast. Therefore, research efforts have begun to explore new low index cladding options such as transparent conductive oxides and lattice-matched quaternary alloys.

In this work, I explore an alternative cladding design using etched nano-structures to lower the effective refractive index and create a high index-contrast top cladding layer. I present detailed simulations, design, and fabrication of blue (435.5nm, 451nm) laser diodes. I also consider the effect of sub-surface dry etch damage which can destroy the light emitting active region. While prior work on light emitting GaN nano-structures required GaN regrowths or recovery anneals, I have developed a low-damage dry etch that avoids the etch damage issue. The resulting process is a new fabrication method for surface etched nano-structures in GaN light emitting devices.