Parallel System Architectures
Parallel System Architectures
Lecturer: Andy D.
Pimentel
Period: Semester one, Block one
Schedule and locations: See Datanose schedule (Hoorcollege = Lecture, Computerpracticum = Lab session)
Lab assistants: Simon Polstra
and Jun Xiao
The objective
of this course is to understand the principles and design of
modern microprocessor systems and how such processors can be
aggregated, using interconnection networks, into
parallel/distributed computing systems. Here, insight will be
provided in both high-performance, general-purpose computer
systems and more constrained embedded computer systems.
The course builds on
a basic knowledge of microprocessor architecture. It develops
this with and emphasis on instruction-level concurrency in
microprocessor design and concurrency in both memory systems and
parallel/distributed computing systems. The topics that are
covered include superscalar and VLIW processor architectures,
instruction- and thread-level parallelism, memory hierarchy,
distributed- and shared-memory parallel computers,
interconnection networks and new architecture trends.
Programming
skills in C/C++, computer organization.