Seminar • Systems and Networking • Taming Model Incongruities in Networked Systems

Please note: This seminar will take place in DC 1304.

Nirav Atre, PhD candidate
Computer Science Department, Carnegie Mellon University

We reason about computer systems via models of their behavior — whether implicit mental models, or explicit mathematical models. These models are the linchpins of our decision-making ability, e.g., in formulating service-level agreements (SLAs), or tendering performance claims. Unfortunately, a growing disconnect between how systems are modeled and how they are actually deployed has engendered a class of problems I call model incongruity: circumstances where a model’s prediction deviates significantly from real-world behavior. Model incongruities are highly pervasive in modern systems, resulting in expensive performance anomalies, scalability bottlenecks, and security vulnerabilities.

My research revolves around systematically addressing model incongruities in a broad range of networked systems, two of which I will highlight in this talk. First, I will describe “delayed hits”, an incongruity arising in high-performance caching systems which (a) breaks the textbook caching principle that maximizing cache hit-rate also minimizes latency, and (b) causes every existing caching algorithm to make latency-suboptimal decisions; in this context, I will introduce Minimum-AggregateDelay (MAD), a turnkey augmentation to existing algorithms that makes them aware of delayed hits, yielding 5–35% lower request latencies. Second, I will describe “algorithmic complexity attacks” (ACAs), a highly potent class of Denial-of-Service attacks arising from transient workload incongruity; in this context, I will introduce SurgeProtector, an adversarial scheduling framework that provably protects network dataplanes against ACAs, resulting in 90–99% reduction in harm for the same volume of attack traffic. Overall, this approach illustrates how treating system models as first-class citizens ultimately enables building more performant, scalable, and resilient systems, even under uncertainty.

Bio: Nirav Atre is a Ph.D. candidate in the Computer Science Department at Carnegie Mellon University (CMU), where he is advised by Prof. Justine Sherry. Nirav is broadly interested in the research space at the intersection of systems and modeling: building high-performance networked systems (in both hardware and software), and proving useful theoretical properties about their performance, scalability, and security.

For his work, he has been awarded a CyLab Presidential Fellowship, named a French-American Doctoral Exchange (FADEx) Laureate in Cybersecurity, and received Best Paper and Distinguished Artifact awards at OSDI. Prior to CMU, Nirav received his B.A.Sc. in Computer Engineering from the University of Toronto, Canada, in 2018.