Skip Navigation CECS Home | ANU Home | Search ANU

COMP8320:
Multicore Computing: Principles and Practice:
Semester 2 2011

Course staff: Dr Peter Strazdins (coordinator and lecturer)

Software engineers who do not understand parallel [multicore] processing will become obsolete! -- Professor Rudolph Eigenmann, keynote address at the ISPA'06 conference

---

Update for 2011:

• We will be getting a state-of-the-art Intel Single Chip Cloud Computer (SCC)! This is an experimental 48-core chip whose on-chip network is without cache coherency!
• We will not run the Small Team Research Projects in 2011, with the assessment changed to regular assignments, plus (possibly) an essay on a `special interest' topic.
• Instead, both SCC + Multicore On-chip Networks and GPUs will be added as regular course `modules' (2 hour lecture, tute/labs + assignment work).

---

General Information The course's formal requisite is enrolment in the Master of Computing. However, one-off enrollments may be made by people with the required assumed knowledge. Also students taking other degrees, including 4th year undergraduates, may seek to enrol through special permission (contact the course co-ordinator). Assumed knowledge is equivalent to having done the equivalent of an introductory course on computer architecture, a course on concurrency, and intermediate programming and data structure courses. Please see the StudyAt entry for Course Description and Learning Outcomes. Lecture times and venue: see the ANU timetable. A total of 10 two-hour lectures will be given, see the course activity schedule for more details. A Course Administration Handout will be handed out in week 1. There is a Discussion forum which will be (semi-) regularly monitored by the course staff. The course's wattle page will be used for DLD recodings. Computing facilities: we will be using an UltraSPARC T2 (also known as the Niagara-2), T5120 model, which was generously donated by Sun Microsystems in 2008. The T2's control domain, called mavericks, is hosted in the Computer System's group research subnet; a guest domain, called wallaman, is exported to the CSIT student subnet (try ssh wallaman). Later in the semester, we will also be using GPUs on the NCI National Facility, and a state-of-the-art (indeed experimental) Intel Single-chip Cloud Computer. Information for Other Multicore Computing Course Instructors Click here.

Copyright | Disclaimer | Privacy | Contact ANU