Research interests
Professor Samer Al-Kiswany’s research spans a number of areas in networking, operating systems, distributed systems, and data management and processing engines.
Modern large-scale software systems face two tectonic waves of fundamental changes. First, a hardware wave: Recent dramatic increase in the rate of technological innovation introduced fundamental changes to core infrastructure components including software-defined networks, object-based disks, non-volatile memory, and specialized accelerators. Second, an application wave: Lower cost and the elasticity of the cloud have attracted classical applications and enabled new application domains, such as the proliferation of cloud-backed mobile apps, the introduction of massive machine learning and artificial intelligence applications, and the expansion of high-impact science workflows.
Yet, the current cloud system software stack has been designed in an era of desktop and server applications and adheres to decades-old design principles that are incongruous in today’s data centres, resulting in a software stack that is inadequate to capitalize on the hardware evolution or to efficiently support new applications.
Professor Al-Kiswany’s work focuses on building the next generation of software systems that efficiently leverage recent hardware changes in the infrastructure, and leverage application-domain specific optimizations, to bring significant performance, efficiency, and scalability gains. In recent projects, Professor Al-Kiswany’s produced a new data storage protocol that can leverage recent networking and accelerator technology to significantly accelerate data storage systems operations.
Degrees and awards
- B.Sc. (Jordan University of Science and Technology), MSc, PhD (British Columbia)
- NSERC Postdoctoral Fellowship (2013–2015)
- Killam Doctoral Fellowship (2009–2012)
- BC Industrial Innovation PhD Fellowship (2007–2010)
Representative publications
Mohammed Alfatafta, Basil Alkhatib, Ahmed Alquraan, Samer Al-Kiswany. ”Toward a Generic Fault Tolerance Technique for Partial Network Partitioning.”
USENIX Symposium on Operating Systems Design and Implementation (OSDI), 2020.
Hatem Takruri, Ibrahim Kettaneh, Ahmed Alquraan, Samer Al-Kiswany. ”FLAIR: Accelerating Reads with Consistency-Aware Network Routing.” USENIX Symposium on Networked Systems Design and Implementation (NSDI), Feb. 2020.
Ahmed Alquraan, Hatem Takruri, Mohammed Alfatafta, Samer Al-Kiswany. ”An Analysis of Network-Partitioning Failures in Cloud Systems.” USENIX Symposium on Operating Systems Design and Implementation (OSDI), Oct. 2018.
