Extreme Computing is about making the most of a computing cluster, be it processing huge data sets or serving high-volume websites. The course covers high-performance computing at several levels of the stack:
- Big data problems, the memory hierarchy, latency requirements, performance considerations, and economics
- Processing large amounts of data, sharding, streaming, query processing, and randomised algorithms
- Distributed file systems, fault tolerance, replication, job scheduling, and MapReduce/Hadoop
We aim to make students aware of topics using large (> 1000 node) clusters, but not be tied to particular vendors whereever possible. If you want to understand how companies like Amazon, Facebook, Google, Microsoft, and Twitter work, then this is the course for you.
Lectures are Mondays and Thursdays 17:10-18:00 in George Square Lecture Theatre.
start week 3.
Who to Contact
- Almost Everything. Including all questions about the labs or assignments outside lab sessions.
- Informatics Teaching Organisation
- Assignment extensions and lab time changes
- Computing Support
- Lab and cluster logins (DICE account)
- Harsh marking
- The Original Marker
- Clear marker error (i.e. failed to add marks correctly, missed a solution). Marks can go up or down.
- Marker error appeals. Include e-mail from the original marker. Marks can go up or down.
Lecturer: Kenneth Heafield
Office: 4.21, Informatics Forum, 10 Crichton Street
E-mail: exc at kheafield.com (but please use Piazza)
Students should be familiar with the Unix command line and ssh. Information systems provides a primer on the command line
We do not require a particular programming language. Examples are mostly in Python and Java with occasional C++.
The is a Level-11 course, normally taken in Year 4. It is open
to all 4th year, MInf, and MSc students in Informatics.
Announcements regarding the course will be posted to the course mailing list
students taking the course are automatically subscribed to this
list. Previous postings can be accessed using the mailing