Abstract

Datacenter systems and I/O devices now run at single-digit microsecond latencies, requiring ns-scale operating systems. Traditional kernel-based operating systems impose an unaffordable overhead, so recent kernel-bypass OSes and libraries eliminate the OS kernel from the I/O datapath. However, none of these systems offer a general-purpose datapath OS replacement that meet the needs of µs-scale systems. We propose Demikernel, a flexible datapath OS and architecture designed for heterogenous kernel-bypass devices and µs-scale datacenter systems. We build two prototype Demikernel OSes and show that minimal effort is needed to port existing µs-scale systems. Once ported, Demikernel lets applications run across heterogenous kernel-bypass devices with ns-scale overheads and no code changes.

Bio: I am a researcher at Microsoft Research. My research focuses on datacenter operating systems and distributed systems. I am 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. My most recent work has been on the Demikernel datapath OS and related projects. A bio and talk abstract for my most recent work is available. I completed my PhD in 2017 at the University of Washington, where my research focused on distributed systems that span mobile devices and cloud servers. My thesis work received the ACM SIGOPS Dennis Ritchie doctoral dissertation award and the UW Allen School William Chan doctoral dissertation award. Before my PhD, I received my S.B. and M.Eng. from MIT and worked for 3 years in the virtual machine monitor group at VMware.