SHF: Small: Computer Architecture for Scripting Languages

University Of Illinois At Urbana-Champaign, Urbana IL

Investigators

Abstract

The goal of this work is to speed-up the execution of the most popular programming languages of the internet, namely Scripting Languages such as JavaScript and Python. These languages are easy to use, because they do not require declaring all the variables. However, this makes them very hard to compile efficiently. In this work, the principal investigator will design processor hardware that is specifically equipped to help speed-up these languages. Given the importance of these languages for web development, effective techniques to speed-up their execution are fundamental for the continued development of the web infrastructure. Such development is crucial to our society's efficiency, competitiveness, and wellbeing. The principal investigator will focus on several techniques. One of them is using speculation support to execute the most likely execution path of the codes, and undo the execution if another path was taken instead. The other technique is prediction, to predict which parts of the code will be executed. Finally, the third technique is reuse, where the processor will keep enough state in its structures to reuse it across invocations of the same functions.