Resolving Feature Convolution in Software Systems Infrastructures

Speaker:	Charles ZHANG
		University of Toronto

Title:		"Resolving Feature Convolution in Software Systems
		 Infrastructures"

Date:		Monday, 2 April 2007

Time:		4:00pm - 5:00pm

Venue:		Lecture Theatre F
		(Leung Yat Sing Lecture Theatre, near lift nos. 25/26)
		HKUST

Abstract:

Large software systems infrastructures for distributed systems
increasingly suffer from complex development and suboptimal performance.
Our observations show that this problem is mostly caused by many
inherently non-modular features only applicable in specific application
contexts. In this talk I present our aspect oriented solutions to this
problem by first describing algorithms that automatically discover these
features in the sources of large software systems. I then focus on the
just-in-time architectural synthesis paradigm which enables software
infrastructures to structurally adapt to how they are being used. I
present the quantitative evaluations using both system benchmarks and
software engineering metrics and show that our approaches benefit both the
development and computational efficiency of software infrastructures.


********************
Biography:

Charles Zhang is a PhD candidate in the Middleware Systems Research Group
at the University of Toronto. He is primarily interested in studying
software architectural methodologies in the context of software systems
infrastructures and system software. The related research effort includes
both the program analysis of legacy software systems and the construction
of novel architectures for software systems infrastructures that are
highly versatile and customizable. He has published extensively at premium
conferences and journals such as IEEE TPDS, OOPSLA, ACM/USENIX MIDDLEWARE,
and AOSD. He is also a two-time IBM PhD fellowship winner.

Charles obtained both his B.Sc. with honors and Master degrees also from
University of Toronto. Prior to his science endeavor, he spent a year in
Beijing Normal University studying history. Before his graduate study, he
worked as a software engineer at Motorola and a Silicon Valley startup.