University home »
Faculty of Science »
Department of Computer Science »
Courses » COMPSCI 703 S1 C » Lectures »
Computer Science
Lectures: COMPSCI 703 Semester 1, City Campus
Lecture schedule, notes, and other handouts.
- Week 1: 2-5March
- Week 2: 9-12Mar
- Week 3: 16-19Mar
- Tuesday: ILP: Hardware-based Speculation (H&P Sections 2.6-2.8) (slides)
- Thursday: Caches & Memory Systems (slides)
- Friday:
Caches & Memory Systems
(slides)
Kozyrakis Lecture Notes
-
Recommended readings for Memory Systems:
- Wikipedia: Cache Memory
- Hill & Smith, "Evaluating associativity in caches" IEEE Trans. on Computers, 38(12), December 1989, pp. 1612-1630.
- Jon Stokes, Understanding CPU caching and performance."
- Week 4: 23-26Mar
- Tuesday: Programming Parallel Computers 1 (slides)
- Thursday: Programming Parallel Computers 2
(Message-passing slides) - Friday: Programming Parallel Computers 3 (Pthreads & Locks slides, fibpt.c)
-
Readings for Week 4:
- How to Survive the Multicore Software Revolution (or at Least Survive the Hype) , (pdf)
- OpenMP Tutorial
- Pthreads Tutorials
- MPI Tutorial
- Topics for revision: The C Programming Language -- Kernighan and Ritchie (Amazon, Google, Library); Processes and Threads
-
Additional Reading:
Ars Technica Snow Leopard Review (pages 11-12): Concurrency in the real world: a prelude , Grand Central Dispatch
-
Assignment 1
- Programming in OpenMP Due Wednesday 21 April
- Week 5: 30Mar-1Apr
- Week 6: 20Apr-23Apr
- Week 7: 27Apr-30Apr
- Week 8: 4May-7May
- Tuesday: Transactional Memory (slides)
- Thursday: In-class test
- Friday: NO CLASS (Graduation)
- Week 9: 11May-14May
- Tuesday: Speculative Lock Elision (slides)
- R. Rajwar & J.R. Goodman, Speculative Lock Elision: enabling highly concurrent multithreaded execution, MICRO-34, December 2001, pp. 294-305.
- R. Rajwar & J.R. Goodman, Transactional execution: toward reliable, high-performance multithreading, IEEE Micro, 23(6), pp. 117-125, Nov/Dec 2003.
- Thursday: Hardware Transactional Memory (slides)
- D Dice, Y Lev, M Moir, D Nussbaum, “Early experience with a commercial hardware transactional memory implementation,” ASPLOS09, pp. 157-168, (March 2009).
- AMD Advanced Synchronization Facility: Proposed Architectural Specfication, Advanced Micro Devices, Publication #45432(revision 2.1), March 2009.
- Friday: Unbounded Hardware Transactional Memory
Sources: - Larus & Rajwar, Chap. 4.4 & 4.5
- Week 10: 18May-21May
Assignment 2 (Project) Due Friday, 4June
- Tuesday: Software Transactional Memory: Non-blocking algorithms
Sources: - Paul E. McKenney: What is RCU, Really?
- Wikipedia: Non-blocking Synchronization
- Angela Demke Brown: Avoiding Locks
- Thursday & Friday: Software Transactional Memory
Sources: - M Herlihy, V Luchangco, M Moir, W Scherer III, Software Transactional Memory for Dynamic-Sized Data Structures (DSTM) Twenty-Second ACM Symposium on Principles of Distributed Computing (PODC), 2003, pp. 92-101.
- NOrec: Streamlining STM by Abolishing Ownership Records
- NZTM: Nonblocking Zero-indirection Transactional Memory (NZSTM)
- Software Transactional Memory Should Not Be Obstruction-Free
- Tuesday: Software Transactional Memory: Non-blocking algorithms
- Week 11: 25May-28May
- Tuesday: Hybrid Transactional Memory
- Larus & Rajwar: 4.6
- Thursday: Historic Parallel Computers
Sources: - M.J. Flynn, "Very high-speed computing systems", IEEE Proceedings, 54(12), pp. 1901-1909, 1966.
- W.J. Bouknight, S.A. Denenberg, D.E. McIntyre, J.M. Randall, A.H.Sameh, and D.L.Slotnick, “The ILLIAC IV system”, IEEE Proceedings, 60(4), pp.369-388, April 1972.
- R.M. Russell, “The CRAY-1 computer system,” CACM, 21(1), pp. 63-72, 1978.
- Friday: Current Parallel Computers
Sources: - J. Nickolls & W.J. Dally, "The GPU Computing Era," IEEE Micro, 30(2), pp. 56-69, Mar/Apr 2010
- L. Seiler, D. Carmean, E. Sprangle, T. Forsyth, M. Abrash, P. Dubey S. Junkins, A. Lake, J. Sugerman, R. Cavin, R. Espasa, E. Grochowski, T. Juan & P. Hanrahan, "Larrabee: a many-core x86 architecture for visual computing," IEEE Micro, 30(2), pp. 56-69, Mar/Apr 2010
- "The Cell architecture", Innovation Matters, IBM Corp.
- NVIDIA CUDA Programming Guide
- Tuesday: NO CLASS!
- Thursday: The Future of Parallel Computers
- Friday: CS703 review
(slides)
Project report due today!
Model Answers for test
-
Related Programmes
