Multi-Core Processor Systems

Course catalogue number	5284MCPS6Y
Credits	6 EC
Language of instruction	English
Time period(s)	Sem. 1 Sem. 2 See also
College/graduate	Graduate School of Informatics
Lecturer(s)	dr. A.D. Pimentel (co-ordinator)
Contact	Education Desk FNWI Science Park 904 B1 020 5257100 servicedesk-esc-science@uva.nl
Is part of	Master's Computer Science (joint degree)
Add to course registration

Objectives

• To understand the principles and design of modern multi-core microprocessors and and their microarchitectures.
• To understand how such processors can be deployed in systems targeted for different application fields, such a as embedded / edge systems-on-a-chip or parallel/distributed computing systems.
• Building a simulation model of a cache-coherent multi-core system, with which cache behavior in such systems can be analyzed and evaluated.
• To develop an understanding of how such multi-core processors can be programmed.

Contents

This course builds on a basic knowledge of microprocessor architecture. It develops this with and emphasis on instruction-level and thread-level concurrency in (multi-core) microprocessor design as well as concurrency in memory 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.

The course also provides an introduction into programming models and techniques for multi-core processors found in everyday laptops to large-scale server systems. Here, the focus will be on multithreaded programming models for shared address space systems, where we discuss both OpenMP compiler directives and Posix threads.

Recommended prior knowledge

The course requires a solid background in system-oriented programming using the C or C++ programming language, and basic knowledge on computer organization.

Registration

More information about procedures and registration periods can be found at https://student.uva.nl/en/topics/course-registration

Teaching method and contact hours

• Lecture
• Self-study
• Computer lab session/practical training

Study materials

• Literature:
  • John Hennessy and David Patterson, Computer Architecture: a Quantitative Approach, Morgan Kaufmann, 3rd ed. or later.
• Practical training material:
  • Study material for the part on parallel programming will be provided during the course.

Assessment

Practical assignments, theory tests/assignments, review of a scientific paper.