University of Glasgow - School of Computing Science logo

Wing Hang Li

email: w.li.2@research.gla.ac.uk

Picture of Wing Hang Li

Hello, my name is Wing Hang Li. I am currently a PhD student at the School of Computing Science at the University of Glasgow.  I am supervised by Dr. Jeremy Singer and Dr John O'Donnell. I also receive advice from Dr. David White.

My research project is "Characterising and Optimising Non-Java Workloads on a JVM". Java is a platform neutral language that uses an abstract computer called a JVM (Java Virtula Machine) to execute its programs on a variety of platforms. Java programs are compiled into bytecode which can be executed by the JVM. Many other languages like Scala, Clojure, JRuby and Jython are now also targetting the JVM. Their own compilers can produce bytecode like Java's but there are differences due to the features of each language. For instance Java is a statically typed language but Clojure, Jruby and Jython are dynamically typed. Java is imperative in style while Scala and Clojure are functional languages. These differences are the reason that non-Java bytecode tends to be slower than Java's when performing the same tasks. However support for non-Java programming languages is being added to JVMs with the JSR292 adding support for dynamically typed programming languages with a new invokedynamic bytecode.

My project will look at the differences in bytecode between Java and non-Java programming languages. I will investigate whether it is possible to identify the programming language from the bytecode it produces and what differences there are in the bytecode of benchmarks written in different languages but performing the same work. Any differences found would be targets for optimisation.

Files related to my JVM languages study are available at http://www.dcs.gla.ac.uk/~wingli/jvm_language_study/.

My previous studies include:

2003-2007 University of the West of Scotland
BSc Computer Networking

2010-2011 University of Glasgow
MSc Information Technology