cse290arch: Architecture Seminar

EBU3B 4140
Wednesday, 11:00-12:00.
Spring, 2008

Course Description

The primary objective of the architecture seminar is to keep us abreast of interesting developments in architecture research with relatively little effort.

Each week everyone reads a paper and one or two students present it. Presentations are low-key:

Questions? email swanson at cs.ucsd.edu

Schedule

Date Presenter Paper
April 9 Steve Paper assignments
April 16 Jack Exploiting access semantics and program behavior to reduce snoop power in chip multiprocessors
April 23 Rick Archipelago: trading address space for reliability and security
April 30 DJ Feedback-driven threading: power-efficient and high-performance execution of multi-threaded workloads on CMPs
May 7 Ganesh Merge: a programming model for heterogeneous multi-core systems
May 14 Anshuman Understanding and visualizing full systems with data flow tomography
May 21 MD Optimistic parallelism benefits from data partitioning
May 28 Leo PICSEL: measuring user-perceived performance to control dynamic frequency scaling
June 4 Laura Better bug reporting with better privacy

Papers

1
``Exploiting access semantics and program behavior to reduce snoop power in chip multiprocessors,'', C. S. Ballapuram, A. Sharif, and H.-H. S. Lee in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 60-69, ACM, 2008, link.

2
``PICSEL: measuring user-perceived performance to control dynamic frequency scaling,'', A. Mallik, J. Cosgrove, R. P. Dick, G. Memik, and P. Dinda in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 70-79, ACM, 2008, link.

3
``Archipelago: trading address space for reliability and security,'', V. B. Lvin, G. Novark, E. D. Berger, and B. G. Zorn in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 115-124, ACM, 2008, link.

4
``Understanding and visualizing full systems with data flow tomography,'', S. Mysore, B. Mazloom, B. Agrawal, and T. Sherwood in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 211-221, ACM, 2008, link.

5
``Xoc, an extension-oriented compiler for systems programming,'', R. Cox, T. Bergan, A. T. Clements, F. Kaashoek, and E. Kohler in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 244-254, ACM, 2008, link.

6
``Merge: a programming model for heterogeneous multi-core systems,'', M. D. Linderman, J. D. Collins, H. Wang, and T. H. Meng in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 287-296, ACM, 2008, link.

7
``Better bug reporting with better privacy,'', M. Castro, M. Costa, and J.-P. Martin in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 319-328, ACM, 2008, link.

8
``Learning from mistakes: a comprehensive study on real world concurrency bug characteristics,'', S. Lu, S. Park, E. Seo, and Y. Zhou in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 329-339, ACM, 2008, link.

9
``Optimistic parallelism benefits from data partitioning,'', M. Kulkarni, K. Pingali, G. Ramanarayanan, B. Walter, K. Bala, and L. P. Chew in ASPLOS XIII: Proceedings of the 13th international conference on Architectural support for programming languages and operating systems, (New York, NY, USA), pp. 233-243, ACM, 2008, link.

10
M. A. Suleman, M. K. Qureshi, and Y. N. Patt, ``Feedback-driven threading: power-efficient and high-performance execution of multi-threaded workloads on CMPs,'' SIGARCH Comput. Archit. News, vol. 36, no. 1, pp. 277-286, 2008.