Lecturer(s)
|
|
Course content
|
Content of lectures : 1. Introduction, basic terms - cloud and grid computing, grid architectures, cloud federations, IaaS, PaaS 2. Virtualizers, principles and technologies 3. Systems for cloud creation and management - CloudStack, OpenStack and OpenNebula 4. Proprietary in-house solution prezentation - system CMU 5. Network technologies in supercomputing, Software Defined Networking, dynamic load balancing 6. Data storages for extreme performance, high powerfull file systems 7. Massive parallelization methods, architecture Intel LaraBee/Xeon Phi 8. Basics on Xeon Phi programming 9. More advanced methodd in Xeon Phi programming I 10. More advanced methodd in Xeon Phi programming II 11. GPU architectures a their using in supercomputing 12. Big Data processing, Apache Hadoop 13. NoSQL databases for Big Data, geographically oriented databases and their optimization Content of practicals: Practical exercises in programming Xeon Phi
|
Learning activities and teaching methods
|
Monologic (reading, lecture, briefing)
- Preparation for classes
- 10 hours per semester
- Preparation for credit
- 30 hours per semester
- Preparation for exam
- 40 hours per semester
- Semestral paper
- 30 hours per semester
- Class attendance
- 40 hours per semester
|
Learning outcomes
|
The aim of the course is to acquaint students with the principles of large computer systems.
Basic overview of technologies and concepts used in today's supercomputers.
|
Prerequisites
|
Basic knowledge of computer systems and networks.
|
Assessment methods and criteria
|
Oral examination, Test
Writing a semester test with a success rate of over 50%.
|
Recommended literature
|
-
Jeffers, Reinders, Intel Xeon Phi Coprocessor High Performance Programming, Morgan Kaufmann, 2013, ISBN: 978-0-12-410414-3.
-
Reinders, Jeffers, High Performance Parallelism Pearls, ElSci, 2015, ISBN: 978-0-12-802118-7.
|