Abstract

Abstract Recent advances in networking technology have significantly increased network configurability, programmability, and flexibility. These advances have introduced programmability to all network components—from programmable switches and network cards to the deployment of configurable software switches on all nodes in an infrastructure.

This increase in network “softwarization” is a double-edged sword. On the one hand, network programmability facilitates the building of line-rate application-specific packet processing logic. This in turn enables building in-network system-specific optimizations, such as optimized distributed query processing, scheduling, and load balancing. In this talk, I will present FlairKV (NSDI ‘20), a key-value storage system that leverages programmable switches to significantly accelerate read operations.

On the other hand, increased network softwarization (perhaps not surprisingly) contributes to an increase in the frequency and complexity of network failures. Can modern systems tolerate these faults, and can we build a generic fault tolerance technique for them? In this talk, I will present NEAT (OSDI ‘18), a comprehensive study of the impact of network partitioning failures on modern systems. Moreover, we recently identified a peculiar type of network fault called partial partitioning. I will present NIFTY (OSDI ‘20), a generic fault tolerance technique for partial partitions.

Bio: Samer Al-Kiswany is an assistant professor at David Cheriton School of Computer Science at the University of Waterloo, Canada. His research interest is in distributed systems, networking, and data management and processing engines. In particular, his work focuses on reconsidering systems design in light of recent changes in cloud applications and platforms. Samer received his PhD from the University of British Columbia in 2013. After his PhD, he joined the University of Wisconsin–Madison as a postdoctoral fellow. Dr. Al-Kiswany is the recipient of ten national and international awards, including the Killam Doctoral Fellowship, the NSERC Postdoctoral Fellowship, and the IEEE George Michael HPC Fellowship.