Irene Zhang

Abstract

Kernel-bypass is the classic example of offloading a common workload to hardware. Every application goes through the OS kernel, therefore it makes an excellent target for baking into hardware. While kernel-bypass has been around for a long time, it has never been as urgently needed as now. Unfortunately, as a research community, we've put very little thought into how to make kernel-bypass actually usable for most applications. RDMA, DPDK and previous hardware were all largely designed for specialized, high-performance applications, making them difficult to use and deploy. With Demikernel, our aim was to bring the principles of the OS kernel to kernel-bypass – including high-level APIs and portability – thus making kernel-bypass accessible to all datacenter applications and developers.

Biography

Irene Zhang is researcher at Microsoft Research.
Her research focuses on datacenter operating systems and distributed systems.
Irene Zhang is especially interested in making it easier to build microsecond-scale datacenter systems by leveraging new datacenter hardware technologies, like kernel-bypass, RDMA and programmable devices.
She completed her PhD in 2017 at the University of Washington, where her research focused on distributed systems that span mobile devices and cloud servers. Her thesis work received the ACM SIGOPS Dennis Ritchie doctoral dissertation award and the UW Allen School William Chan doctoral dissertion award. Before her PhD, she received her S.B. and M.Eng. from MIT and worked for 3 years in the virtual machine monitor group at VMware.
Irene was born in Beijing, China but spent most of her time growing up in Columbus, Indiana. Irene and her husband like to cook, travel and occasionally do computer science together.
Her most recent work has been on the Demikernel datapath OS and related projects.
Irene will present Demikernel during this talk.