For the past 35 years or so, I’ve been doing systems research in the general areas of operating systems and distributed systems, with occasional excursions into security, end-user applications, and one physical gadget (the PJB). For historical details, see the biography and publication list on my personal web site (which also contains a large pile of other stuff, mostly unrelated to computer science).

Since mid-2009, I've been working almost exclusively on the Beehive project. Beehive is a new multi-core computer, implemented on an FPGA. It is designed to enable research and teaching at the boundary of computer architecture and low-level systems programming. Beehive users have access to the raw hardware design (in Verilog), a collection of low-level software libraries (in C), and a toolchain with a C compiler. They can modify any or all of these, exploring "what-if" questions by experiment on a 13-core 100 MHz computer. We have made Beehive available to several universities, including MIT and ETH Zurich, with full source and an open license. Beehive is available only to those who have signed our license agreement.

Previously at Microsoft, since coming here in 2001, I’ve worked on Automatic Mutual Exclusion, Dryad, Singularity security, Desktop on a Keychain, LBOW (unpublished) and Penny Black.

My work email is b@m, where b=birrell and m=microsoft.com; my personal email is a@b.org, where a=andrew and b=birrell.