This section contains video lectures and lecture slides. http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures 2013-01-12T07:51:03+05:00 MIT OpenCourseWare http://ocw.mit.edu en-US Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Introduction to course, administrative information. Lecture covering matrix multiply as a case study, including matrix representation, performance counters, instruction level optimizations, and parallel execution.</p> <p><strong>Speakers:</strong> Charles Leiserson, Saman Amarasinghe</p>Keywords: matrix multiply, profiling, matrix representation, performance counters, instruction level optimizations, parallel execution<br><br>Thumbnail - <a href= http://img.youtube.com/vi/JzpkXLH9zLQ/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec01_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-1-matrix-multiply/id481759887?i=109649113>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/JzpkXLH9zLQ>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-1-matrix-multiply-a-case-study Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US matrix multiply profiling matrix representation performance counters instruction level optimizations parallel execution MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering bit hacks. Examples include swap, population count, and the n-queens problem, used to highlight backtracking search and bitvector representation.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: swap, rounding up, population count, n-queens, backtracking, bitvector representation<br><br>Thumbnail - <a href= http://img.youtube.com/vi/xc9DDSbf0NQ/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec02_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-2-bit-hacks/id481759887?i=109649104>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/xc9DDSbf0NQ>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-2-bit-hacks Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US swap rounding up population count n-queens backtracking bitvector representation MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Basic performance engineering. Bentley's rules (modifying data, modifying code) and the traveling salesman problem.</p> <p><strong>Speaker:</strong> Saman Amarasinghe</p>Keywords: Bentley's rules, traveling salesman problem, modifying data, modifying code, loop rules, logic rules, procedure rules, expression rules, parallelism rules<br><br>Thumbnail - <a href= http://img.youtube.com/vi/2fjZhoifOiM/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec03_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-3-basic-performance/id481759887?i=109649110>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/2fjZhoifOiM>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-3-basic-performance-engineering Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US Bentley's rules traveling salesman problem modifying data modifying code loop rules logic rules procedure rules expression rules parallelism rules MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Overview of computer architecture, profiling a program, and a set of example programs. Student performance on project 1 is also discussed.</p> <p><strong>Speakers:</strong> Saman Amarasinghe, John Dong</p>Keywords: computer architecture, profiling, pipelining, SIMD<br><br>Thumbnail - <a href= http://img.youtube.com/vi/8dTMUigqBHM/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec04_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-4-computer-architecture/id481759887?i=109649096>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/8dTMUigqBHM>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-4-computer-architecture-and-performance-engineering Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US computer architecture profiling pipelining SIMD MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Meeting for 6.172 industry mentors. Description of mentorship role, expectations, overview of course and how it fits into EECS curriculum.</p> <p><strong>Speakers:</strong> Saman Amarasinghe, Charles Leiserson, Eirik Bakke</p>Keywords: industry mentor, MITPOSSE, coding style, coding practices<br><br>Thumbnail - <a href= http://img.youtube.com/vi/nukI0huUEiM/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_mentors_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/industry-mentor-mitposse-overview/id481759887?i=109649101>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/nukI0huUEiM>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/industry-mentor-mitposse-overview Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US industry mentor MITPOSSE coding style coding practices MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Theory and background of profiling tools. Two interactive walkthroughs: matrix multiply and branchless sorting.</p> <p><strong>Speakers:</strong> Reid Kleckner, John Dong, Saman Amarasinghe</p>Keywords: profiling tools, matrix multiply, branchless sorting, hotspots, perf<br><br>Thumbnail - <a href= http://img.youtube.com/vi/7a89iFEEpTo/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec05_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-5-performance-engineering/id481759887?i=109649107>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/7a89iFEEpTo>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-5-performance-engineering-with-profiling-tools Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US profiling tools matrix multiply branchless sorting hotspots perf MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering single threaded performance and assembly language. Discussion of project results, emphasizing importance of writing tests first and pair programming.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: test driven development, pair programming, single threaded performance, assembly language, subroutine linkage<br><br>Thumbnail - <a href= http://img.youtube.com/vi/GOlurMtkuWQ/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec06_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-6-c-to-assembler/id481759887?i=109649108>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/GOlurMtkuWQ>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-6-c-to-assembler Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US test driven development pair programming single threaded performance assembly language subroutine linkage MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering memory systems, including cache concepts, access pattern concepts, data structure transformations, and computation transformations.</p> <p><strong>Speaker:</strong> Saman Amarasinghe</p>Keywords: memory systems, cache, memory access pattern, blocking<br><br>Thumbnail - <a href= http://img.youtube.com/vi/l8GqOCN8M-w/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec07_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-7-memory-systems-performance/id481759887?i=109649117>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/l8GqOCN8M-w>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-7-memory-systems-and-performance-engineering Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US memory systems cache memory access pattern blocking MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering cache-efficient algorithms, with tiled and recursive matrix multiplication examples.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: memory hierarchy, ideal cache model, tall cache, tiled matrix multiplication, recursive matrix multiplication, cache misses<br><br>Thumbnail - <a href= http://img.youtube.com/vi/T9LkSKK075M/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec08_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-8-cache-efficient/id481759887?i=109649103>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/T9LkSKK075M>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-8-cache-efficient-algorithms Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US memory hierarchy ideal cache model tall cache tiled matrix multiplication recursive matrix multiplication cache misses MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Discussion of project 2.2 beta, how pair programming can lead to less time debugging, and importance of working well in groups. Lecture covering more cache-efficient algorithms, including a heat diffusion simulation.</p> <p><strong>Speakers:</strong> John Dong, Saman Amarasinghe, Charles Leiserson</p>Keywords: merge sort, k-funnel, funnelsort, multiway merge sort, heat diffusion simulation, stencil<br><br>Thumbnail - <a href= http://img.youtube.com/vi/CxuZZcUKxWM/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec09_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-9-cache-efficient/id481759887?i=109649099>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/CxuZZcUKxWM>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-9-cache-efficient-algorithms-ii Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US merge sort k-funnel funnelsort multiway merge sort heat diffusion simulation stencil MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering dynamic storage allocation, including reference counting, a graph abstraction, and updating pointers.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: stack allocation, heap allocation, garbage collection, reference counting, graph abstraction<br><br>Thumbnail - <a href= http://img.youtube.com/vi/p0bc1f6ULxw/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec10_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-10-dynamic-storage/id481759887?i=109649118>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/p0bc1f6ULxw>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-10-dynamic-storage-allocation Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US stack allocation heap allocation garbage collection reference counting graph abstraction MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering compiler hacks, when to optimize, data-flow analysis and optimizations, and instruction scheduling. Discussion of quiz, including common mistakes.</p> <p><strong>Speakers:</strong> Saman Amarasinghe, Charles Leiserson</p>Keywords: compilers, vectorization, optimization, data flow analysis, instruction scheduling<br><br>Thumbnail - <a href= http://img.youtube.com/vi/HH1k11sdlTc/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec11_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-11-what-compilers/id481759887?i=109649105>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/HH1k11sdlTc>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-11-what-compilers-can-and-cannot-do Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US compilers vectorization optimization data flow analysis instruction scheduling MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Discussion of project 3 beta. Lecture covering multicore programming, including shared-memory hardware, concurrency platforms, and race conditions.</p> <p><strong>Speakers:</strong> Charles Leiserson, John Dong</p>Keywords: multicore programming, shared memory, concurrency, pthreads, OpenMP, Cilk++, race conditions<br><br>Thumbnail - <a href= http://img.youtube.com/vi/_40zvhkmtM8/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec12_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-12-multicore-programming/id481759887?i=109649109>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/_40zvhkmtM8>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-12-multicore-programming Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US multicore programming shared memory concurrency pthreads OpenMP Cilk++ race conditions MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering parallelism, scheduling theory, the Cilk++ runtime system, and Cilk chess programs.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: Cilk++ runtime system, scheduling theory, greedy scheduling, Amdahl's Law, Cilkchess<br><br>Thumbnail - <a href= http://img.youtube.com/vi/UUKTIhxznF0/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec13_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-13-parallelism-performance/id481759887?i=109649111>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/UUKTIhxznF0>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-13-parallelism-and-performance Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US Cilk++ runtime system scheduling theory greedy scheduling Amdahl's Law Cilkchess MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering analysis of multithreaded algorithms, including divide-and-conquer recurrences, loop parallelism in Cilk++, and matrix multiplication and merge sort examples.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: multithreaded algorithms, divide and conquer, recurrences, cilk loops, matrix multiplication, merge sort<br><br>Thumbnail - <a href= http://img.youtube.com/vi/F8yY7jri32M/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec14_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-14-analysis-multithreaded/id481759887?i=109649116>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/F8yY7jri32M>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-14-analysis-of-multithreaded-algorithms Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US multithreaded algorithms divide and conquer recurrences cilk loops matrix multiplication merge sort MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering nondeterministic programming, including mutual exclusion, implementation of mutexes, and locking anomalies.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: nondeterministic, mutual exclusion, mutex, deadlock, convoying, contention, critical section, determinacy race, data race, dining philosophers<br><br>Thumbnail - <a href= http://img.youtube.com/vi/gXRmNp4Wgb0/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec15_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-15-nondeterministic/id481759887?i=109649115>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/gXRmNp4Wgb0>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-15-nondeterministic-programming Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US nondeterministic mutual exclusion mutex deadlock convoying contention critical section determinacy race data race dining philosophers MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering synchronizing without locks, including memory consistency, lock-free protocols, the ABA problem, and reducer hyperobjects.</p> <p><strong>Speaker:</strong> Charles Leiserson</p>Keywords: synchronizing, locks, mutual-exclusion problem, Peterson's algorithm, instruction reordering, memory fence, lock-free stack, reducer, monoid<br><br>Thumbnail - <a href= http://img.youtube.com/vi/n_Cqx8KamFA/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec16_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-16-synchronizing-without/id481759887?i=109649254>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/n_Cqx8KamFA>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-16-synchronizing-without-locks Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US synchronizing locks mutual-exclusion problem Peterson's algorithm instruction reordering memory fence lock-free stack reducer monoid MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering the impact of synchronization and memory on parallel performance, using OpenMP instead of Cilk. Topics include granularity of parallelism, true and false sharing, and load balancing.</p> <p><strong>Speaker:</strong> Saman Amarasinghe</p>Keywords: OpenMP, synchronization, memory, parallel performance, true sharing, false sharing, load balancing<br><br>Thumbnail - <a href= http://img.youtube.com/vi/MFmxByf9x88/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec17_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-17-performance-issues/id481759887?i=109649102>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/MFmxByf9x88>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-17-performance-issues-in-parallelization Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US OpenMP synchronization memory parallel performance true sharing false sharing load balancing MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Primer on ray tracing techniques, given to prepare students for the final project. Includes some ray tracing background and a code overview covering classes and high-level execution.</p> <p><strong>Speaker:</strong> Joshua Slocum</p>Keywords: ray tracing, image rendering, ray casting, recursive casting, scattering, photon map, irradiance cache<br><br>Thumbnail - <a href= http://img.youtube.com/vi/c45ieqaUU4k/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec18_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-18-primer-on-ray-tracing/id481759887?i=109649100>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/c45ieqaUU4k>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-18-primer-on-ray-tracing-techniques Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US ray tracing image rendering ray casting recursive casting scattering photon map irradiance cache MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Information about the final project, followed by a guest lecture covering fractal trees, which combine the strengths of B-trees and append-to-file. Analysis of search and insert in a simplified fractal tree.</p> <p><strong>Speaker:</strong> Bradley Kuszmaul</p>Keywords: cache-oblivious, fractal tree, Tokutek, B-tree, lookup, insert, algorithmic performance model, append, fractal tree indexes<br><br>Thumbnail - <a href= http://img.youtube.com/vi/9Rb85cOXTKU/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec19_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-19-how-tokudb-fractal/id481759887?i=109649112>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/9Rb85cOXTKU>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-19-how-tokudb-fractal-tree-indexes-work Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US cache-oblivious fractal tree Tokutek B-tree lookup insert algorithmic performance model append fractal tree indexes MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Lecture covering distributed systems at the cluster, data center, and planet scales. Topics include message passing, examples of the need to scale, Google's programming model, and cloud computing.</p> <p><strong>Speaker:</strong> Saman Amarasinghe</p>Keywords: distributed systems, scaling up, clusters, data centers, shared memory, message passing, web page analysis, scene completion, virtual machines, cloud computing, deduplication<br><br>Thumbnail - <a href= http://img.youtube.com/vi/sGFEsAphyLo/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec20_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-20-distributed-systems/id481759887?i=109649114>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/sGFEsAphyLo>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-20-distributed-systems Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US distributed systems scaling up clusters data centers shared memory message passing web page analysis scene completion virtual machines cloud computing deduplication MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> The first part of the lecture covers parallelism analysis, caches, and synchronization correctness. The second focuses on compiler optimization questions: is the optimization legal, faster, and automatic?</p> <p><strong>Speaker:</strong> John Dong</p>Keywords: parallelism analysis, synchronization correctness, hyperpoint, compiler optimization<br><br>Thumbnail - <a href= http://img.youtube.com/vi/ajXDBKIJsJ8/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec21_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-21-quiz-2-review/id481759887?i=109649097>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/ajXDBKIJsJ8>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-21-quiz-2-review Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US parallelism analysis synchronization correctness hyperpoint compiler optimization MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Guest lecture by Ravi Soundararajan of VMware, covering ten case studies in performance engineering.</p>Keywords: performance, lock percentage, powerCLI, scalable solutions, remote console issues, 32-bit, 64-bit, memory leak, customized profiler, client-server interactions, disk latency, virtual machines, VM activity, power management<br><br>Thumbnail - <a href= http://img.youtube.com/vi/Ahnns47FKHM/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec22_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-22-a-tale-10-bugs/id481759887?i=109649255>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/Ahnns47FKHM>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-22-a-tale-of-10-bugs-performance-engineering-at-vmware Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US performance lock percentage powerCLI scalable solutions remote console issues 32-bit 64-bit memory leak customized profiler client-server interactions disk latency virtual machines VM activity power management MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm <p><strong>Description:</strong> Students vote on which ray-tracer images are accurate enough to be included in the competition, and these entries are tested against each other for speed. Students also provide feedback on the course.</p>Keywords: ray tracer, final competition, final derby, performance engineering, recitations, tools<br><br>Thumbnail - <a href= http://img.youtube.com/vi/dvPJj-5X_uU/default.jpg>JPG (YouTube)</a><br>Video - download: <a href= http://www.archive.org/download/MIT6.172F10/MIT6_172_F10_lec23_300k.mp4>Internet Archive (MP4)</a><br>Video - download: <a href= http://itunes.apple.com/us/itunes-u/lecture-23-final-competition/id481759887?i=109649106>iTunes U (MP4)</a><br>Video - stream: <a href= http://www.youtube.com/v/dvPJj-5X_uU>YouTube </a><br><br><a href= 'http://ocw.mit.edu/terms/'>(CC BY-NC-SA)</a><br><br> http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-172-performance-engineering-of-software-systems-fall-2010/video-lectures/lecture-23-final-competition-and-student-feedback Amarasinghe, Saman Leiserson, Charles 2011-12-22T16:54:49+05:00 en-US ray tracer final competition final derby performance engineering recitations tools MIT OpenCourseWare http://ocw.mit.edu Content within individual OCW courses is (c) by the individual authors unless otherwise noted. MIT OpenCourseWare materials are licensed by the Massachusetts Institute of Technology under a Creative Commons License (Attribution-NonCommercial-ShareAlike). For further information see http://ocw.mit.edu/terms/index.htm