skip to primary navigationskip to content

Department of Computer Science and Technology

Wheeler Lectures


Wheeler Lectures

A History of Virtualization

The fifth annual Wheeler lecture was given at the Computer Laboratory on Wednesday 25th May, 2016. The speaker was Andrew Herbert who gave an overview of virtualisation techniques in operating systems. The lecture was preceded by a series of ‘minute madness’ talks on current research themes.


Computer Laboratory folklore attributes the theory that “any problem in computer science can be solved by another level of indirection” to David Wheeler and indeed the theory can be seen at work in Wheeler’s initial orders and library subroutine system for EDSAC. In computer operating systems indirection is normally associated with some form of “virtualisation”. With roots in early operating system designs of the 1960s, virtualisation has become a key element of modern operating systems structure, especially in cloud computing. The lecture will explore the origins and evolution of virtualisation and the impact on modern computer hardware and operating systems architecture.

Andrew Herbert joined the Computer laboratory in 1975 as a PhD student working with Maurice Wilkes and Roger Needham on the Cambridge CAP Computer. Following subsequent work with Needham developing the Cambridge Distributed System, Andrew left the Laboratory for a career in industry culminating in becoming the Chairman of the Microsoft Research laboratories across Europe, the Middle East and Africa. Having retired in 2010 he is currently managing a project to build a reconstruction of EDSAC at the National Museum of Computing on Bletchley Park. He hopes to have the replica operational this year.

This talk is an extended version of an invited talk given at the ACM Symposium on Operating Systems Principles “History Day” last year.

Minute madness talks

Preceeding the Wheeler lecture were a series of one minute talks showcasing current research themes at the Computer Laboratory. These talks began at 14:30.


The list of talks in order of presentation was as follows:

  • Sam Ainsworth, Programmable Prefetching
  • Gianluca Ascolani, Cancer stemness and heterogeneity
  • Nikilesh Balakrishnan, Data Provenance
  • Kris Cao, Can computers tell stories?
  • Stephen Clark, Google does not know what a transitive verb is
  • Jon Crowcroft, IoT: what could possibly go wrong?
  • John Daugman, Retrieving vaccination and healthcare records in developing countries using iris recognition
  • Lawrence Esswood, Control Flow Integrity For Elf
  • Anthony Fox, Formal Verification with Instruction Set Architectures
  • Zafar Gilani, Stweeler: A framework for Twitter Bot Analysis
  • Maria Gorinova, Data Wrangling in Healthcare
  • Hatice Gunes, Affective and Social Signal Analysis for HCI & HRI
  • Rob Harle, Context Awareness and Sensor Fusion for Mobiles and Wearables
  • Desi Hristova, Geo-Social Networks
  • Alice Hutchings, Stressed out? Denial of service attacks from the providers’ perspective
  • Tim Jones, Program Parallelisation
  • Martin Kleppmann, TRUE DATA: Placing a bit less trust in the cloud
  • Ekaterina Kochmar, When writing matters
  • Wenda Li, Non-linear first order formulas for real numbers in Isabelle/HOL
  • David Llewellyn-Jones, Pico: liberating humanity from passwords
  • Meredydd Luff and Ian Davies, Anvil: Web Apps Made Simple
  • Anil Madhavepeddy, Unikernels
  • Rafal Mantiuk, Perceptual rendering for future displays
  • Russel Moore, Multistage Speech Segmenter (for Learner Speech)
  • Simon Moore, CHERI: Secure Hardware
  • Annalisa Occhipinti, Using Computer Science to inspire young generations
  • Larry Paulson, Mathematics and the Computer
  • Peter Robinson, Computing with Emotions
  • Advait Sarkar, How to make advanced analytics usable by non-experts?
  • Frank Stajano, Making cybersecurity fun
  • Daniel Thomas, Cambridge cybercrime centre
  • Ivo Timoteo, Learning dynamic systems as networks of differential equations
  • Bingjie Wang, Machine Learning for Lemma Recommendation
  • Duo Wang, A new convolutional neural network model for medical image segmentation
  • Robert Watson, CHERI: Secure Software