ACM SIGMETRICS / IFIP PERFORMANCE 2022
June 6-10, 2022
(7:45 pm - 8:45 pm IST / 10:15 am - 11:15 am EDT)
From the 1970s, packet-switched networking has grown alongside areas such as distributed computing, databases, operating systems, computer architecture, fault-tolerant systems and consensus protocols. Together with these fields, it has ushered the cloud computing revolution which drives most of today's retail, banking, finance, healthcare, manufacturing, entertainment and gaming activities.
Whereas mathematical methods and models have had little to no impact in the analysis of distributed systems or in computer architecture, they have been unusually − unreasonably, one might argue − effective in the performance analysis of packet-switched networks. Furthermore, in contrast to the highly successful LQG (Linear, Quadratic, Gaussian) theory in stochastic control, models of networks have proven to be robust to cost-function and distributional assumptions, evolving well with time and application context.
In this talk, I will speculate on the reasons for this success and survey a few of my favorite models and methods of network performance. I will describe some recent work we are doing on "timeliness and time-sensitive systems." Specifically, I will describe a high-precision, software-based network clock synchronization system and show how it can be used to build: (1) a "fair" financial exchange on top of jittery public clouds, and (2) a new congestion control system which can convert commodity packet-switched interconnects into "zero-drop networks;" in the latter case, mathematical modeling continues to enjoy success.
Balaji Prabhakar is VMWare Founders Professor of Computer Science and a faculty member in the Departments of Electrical Engineering and Computer Science, and, by courtesy, in the Graduate School of Business at Stanford University. His research interests are in computer networks; notably, in Data Center Networks and Cloud Computing Platforms. His work spans network algorithms, congestion control protocols, and stochastic network theory. He has also worked on Societal Networks, where he has developed "nudge engines" to incentivize commuters to travel in off-peak times so that congestion, fuel and pollution costs are reduced.
Balaji has been a Terman Fellow at Stanford University, and a Fellow of the Alfred P. Sloan Foundation, IEEE and ACM. He has received the CAREER award from the U.S. National Science Foundation, the Erlang Prize, the Rollo Davidson Prize, and delivered the Lunteren Lectures. He is the recipient of the inaugural IEEE Innovation in Societal Infrastructure Award which recognizes "significant technological achievements and contributions to the establishment, development and proliferation of innovative societal infrastructure systems." He has received the IEEE Koji Kobayashi Award for his work on Computer Communications. He is a co-recipient of several best paper and test of time awards. During 2005−07 he was switch architect at Nuova Systems (acquired by Cisco Systems) where he developed the fabric scheduling and line card algorithms of Cisco's Nexus 5000 family of data center Ethernet switches. In 2011 he co-founded Urban Engines (acquired by Google in 2016) and is currently on leave at Clockwork.io where he is co-founder and CEO.
Carnegie Mellon University
(6:30 pm - 7:30 pm IST / 9:00 am - 10:00 am EDT)
For many years, the research community separated the study of distributed systems (and corresponding applications) from the study of communication networks. This paradigm is starting to change; researchers have found that communication networks can significantly impact application-level properties of distributed systems, including performance, security, privacy, and fairness. For example, such interactions arise in the analysis of epidemics and the design of blockchain systems. In this talk, we highlight how careful modeling, analysis, and design of communication networks can impact the properties of distributed systems. We illustrate this point through a detailed case study focusing on privacy in blockchain systems, while alluding to other properties like performance, security, and fairness. We conclude with open questions for the community regarding how to design distributed systems that inherently account for the role of networks.
Giulia Fanti is an Assistant Professor of Electrical and Computer Engineering at Carnegie Mellon University. Her research interests span the security, privacy, and efficiency of distributed systems. She is a two-time fellow of the World Economic Forum’s Global Future Council on Cybersecurity and a member of NIST’s Information Security and Privacy Advisory Board. Her work has been recognized with best paper awards, a Sloan Fellowship, an Intel Rising Star Faculty Research Award, and a U.S. Air Force Research Laboratory Young Investigator Grant. She obtained her Ph.D. in EECS from U.C. Berkeley and her B.S. in ECE from Olin College of Engineering.