Title: A Vision for Computer Science - The System Perspective
Speaker: Joseph Sifakis (Verimag Laboratory, France)
Time: 2:30pm, Tuesday March 8
Venue: Lecture room, Lab for Computer Science, Level 3 Building #5, Institute of Software, CAS
Abstract: Spectacular and continuous evolution of Information Science and Technology over the past decades is mainly driven by two factors: 1) exponential growth of the computing
power of hardware components; 2) the ever increasing potential for applications in all the sectors of modern economies. The initial focus on scientific computing has
quickly extended to commercial applications with the advent of mainframes in the 70¡¯s. The convergence between computing and telecommunication systems in the 80¡¯s opened
the way to Internet and the Web. Another seamless revolution breaks out in the 90¡¯s with the advent of embedded technologies. Hundreds of billions of embedded components
hidden in devices and appliances provide services by continuously interacting with physical environments. An important anticipated landmark is the internet of things,
result of the convergence between embedded and internet technologies.
This rapid evolution and expansion demands a growing effort in innovative R&D and raises challenging and hard research problems for Computer Science. There is currently an
increasing gap between our technological capabilities for treating and transmitting information and our know-how in computing systems engineering. Harnessing the power
of manycore chips, safe and secure deployment of global services by composing available features, smart management of resources for enhanced autonomy and availability are
examples of challenges brought up by the onset of new technologies and demand for new applications.
In this talk, I will discuss the evolution of Computer Science and in particular its shift of focus from algorithms and programs to systems. I will advocate for a coherent
scientific foundation of system design and present a vision for its development encompassing three work directions:
Marrying Physicality and Computation: Computation models ignore physical time and resources and are by their nature very different from analytic models used in physical
systems engineering. In order to take into account interaction of computing systems with physical environments they must be enriched and extended with paradigms and methods
from Electrical Engineering and Control Theory.
Component-based Construction: Complex systems are designed by assembling heterogeneous components. Heterogeneity has different sources including a large variety of interaction
mechanisms, synchronous or asynchronous execution and different levels of abstraction. We need theoretical frameworks encompassing meaningful and natural composition of
heterogeneous components which is essential for tractable and productive system design.
Guaranteeing Correctness: System design flows are empirical and lack theoretical foundations and methods guaranteeing that the designed systems meet their requirements.
Currently correctness of complex systems is ensured mainly by a posteriori checking which suffers from well-known complexity limitations and incurs high development costs.
We need theory and methods the guarantee at design time system correctness. We distinguish two complementary avenues: One is enforcing by control essential properties to adapt
a system¡¯s behavior to uncertain and dynamically changing environments. The other is inferring global properties of composite systems from properties of their components.
I will conclude with general remarks about the nature of Computer Science as a scientific discipline on its own right and advocate for a deeper interaction and cross-fertilization
with other more mature disciplines.
About the speaker:
Prof. Joseph Sifakis shared ACM Turing Award with Prof. E. Clarke and Prof. A. Emerson in 2007, for his pioneering work on Model Checking.