Course Synopsis
A C.S. alum's perspective on what is taught in CS 547:
2005.
This course provides an introduction to the state-of-the-art analytic
modeling techniques that are used in computer system design.
Topics include fundamental performance laws, bottleneck analysis,
basic probability and probability distributions,
network traffic and server workload characterization,
mean value and state transition models, and elementary queueing
theory. These modeling techniques have been used, for example,
to design optimized operating system semaphores & scheduling policies,
optimized Grid applications, and high performance parallel architectures
such as cache coherence protocols, bottleneck-free interconnection
networks, and simple fair bus arbitration protocols. The techniques have
also been used to design high performance Internet protocols such as
near-optimal video streaming protocols and a new near-optimal TCP protocol.
Analytically-guided designs have been adopted by industry because they are
relatively easy to implement and because they have near-optimal
performance. The course will cover applications to illustrate how the
techniques are applied and the insights that are obtained.
