Alvin AuYoung : Research Statement

I am a systems researcher interested in designing, building, deploying and measuring large-scale systems. My primary research interests currently involve the application of economic theory to both fundamental and emerging problems in operating and distributed systems, including, but not limited to resource allocation and congestion pricing. The arrival of large-scale open platforms such as cloud-computing infrastructures and petascale clusters is fueling the emergence of a new class of users and applications with large and diverse resource needs. For example, an application may prefer twice as much memory at the cost of an additional unit of processing power or storage, or a user may prefer to reduce power consumption at the cost of increased throughput. Optimizing for a traditional system-centric metric such as response time or throughput does not capture the diverse, and often conflicting needs of these stakeholders. As a systems and networking researcher, it is clear that our community needs to revisit classic problems in our area and design systems that can satisfy these diverse needs. The challenge for any system to deliver on these goals is to first elicit this “utility” information from its users and then to determine how to use the corresponding information to meet its goals. Collaboration between economics and computer science researchers over the past ten years has lead to many important results that describe how to design economic-inspired mechanisms to address exactly these issues. However, there is very little progress on how to build systems with these mechanisms. The resulting reality is that systems continue to rely on traditional and inadequate designs despite theoretical results demonstrating the potentially significant value added to all stakeholders by designing economics-inspired systems. My work focuses on bridging this widening gap between theory and practice. I argue that systems research can play a critical role in this effort: through an iterative process of deployment, measurement and modeling, we can uncover limitations of proposed theoretical models, and motivate the development of refined models to better fit emerging system designs. Furthermore, we can address the practical deployment challenges such as reducing the burden on usage, and easing the management and maintenance of these systems. Thus far my research has had impact in bringing to the forefront such limitations and how systems can be designed to address these challenges. On a broader scale, my research agenda is to gain a deeper understanding of when and where such economics-inspired theory can be applied to other problems in systems and networking, and perhaps allow us to rethink traditional systems and networking design in order to allow us to build systems to meet our current and emerging needs.