tddd56 multicore and gpu programming lesson 1
play

TDDD56: Multicore and GPU programming Lesson 1 Introduction to - PowerPoint PPT Presentation

Feb 20, 2024 •428 likes •1.61k views

TDDD56: Multicore and GPU programming Lesson 1 Introduction to laboratory work Nicolas Melot nicolas.melot (at) liu.se Linkping university (Sweden) November 4, 2015 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4,

  1. The ABA problem Thread 0 The scenario starts ◮ old → A Thread 0 pops A, preempted before ◮ new → B CAS(head, old=A, new=B) ◮ pool → A → null Thread 1 pops A, succeeds Thread 1 Thread 2 pops B, succeeds ◮ old → C ◮ new → A Thread 1 pushes A, succeeds ◮ pool → null Thread 0 performs Thread 2 CAS(head, old=A, new=B) ◮ old → B The shared stack should be empty, but it ◮ new → C points to B in Thread 2’s recycling bin ◮ pool → B → null shared → B → null Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 17 / 40

  2. The ABA problem shared stack head A null head null head null head B null thread 0 pool thread 1 pool thread 2 pool Figure: The shared stack should be empty, but points to B in Thread 2’s recycling bin Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 18 / 40

  3. Lab 2: Directions Implement a stack and protect it using locks Implement a CAS-based stack ◮ A CAS assembly implementation is provided in the lab skeleton Use pthread synchronization to make several threads to preempt each other in order to play one ABA scenario Use a ABA-free performance test to compare performance of a lock-based and CAS-based concurrent stack Get more details and hints in the lab compendium Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 19 / 40

  4. Lab 3: Parallel sorting Implement or optimize an existing sequential sort implementation Parallelize with shared memory approach (pthread or openMP) Paralleize with Dataflow (Drake) Test your sorting implementation with various situations ◮ Random, ascending, descending or constant input ◮ Small and big input sizes ◮ Other tricky situations you may imagine Built-in sorting functions (qsort(), std::sort()) are forbidden ◮ May rewrite it for better performance Lab demo: describe the important techniques that accelerate your implementation. Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 20 / 40

  5. Base sequential sort pivot = array[size / 2]; for (i = 0; i < size; i++) { if (array[i] < pivot) { left[left size] = array[i]; left size++; } else if (array[i] > pivot) { right[right size] = array[i]; right size++; } else pivot count++; } simple quicksort(left, left size); simple quicksort(right, right size); memcpy(array, left, left size * sizeof ( int )); for (i = left size; i < left size + pivot count; i++) array[i] = pivot; memcpy(array + left size + pivot count, right, right size * sizeof ( int )); Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 21 / 40

  6. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  7. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  8. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  9. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  10. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  11. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  12. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  13. Base sequential sort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 22 / 40

  14. Parallelization opportunities Parallelization opportunities ◮ Recursive calls ◮ Computing pivots ◮ Merging, if necessary Smart solutions challenging to implement ◮ In-place quicksort: false sharing ◮ Parallel sampling/merging: synchronization ◮ Follow the KISS rule Avoid spawning more threads than the computer has cores Use data locality with caches and cache lines Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 23 / 40

  15. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  16. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  17. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  18. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  19. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  20. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  21. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  22. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  23. Simple parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 24 / 40

  24. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  25. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  26. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  27. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  28. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  29. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  30. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Partition and recurse into 3 parts (sample sort) Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  31. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Partition and recurse into 3 parts (sample sort) Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  32. Parallel Quicksort sort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Choose pivot to divide buffer into unequal parts Partition and recurse into 3 parts (sample sort) Makes implementation harder Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 25 / 40

  33. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  34. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  35. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  36. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  37. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  38. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  39. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  40. Mergesort Data index Processing time sequential task Local sort Partition Merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 26 / 40

  41. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  42. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  43. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  44. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  45. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  46. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  47. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  48. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  49. Simple Mergesort Parallelization Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 27 / 40

  50. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  51. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  52. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  53. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  54. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  55. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Partition and recurse into 3 parts and 3-way merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  56. Parallel Mergesort with 3 cores Can only efficiency use power of two number of cores. How to use three cores efficiently? Divide buffer into 2 unequal parts Partition and recurse into 3 parts and 3-way merge Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 28 / 40

  57. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  58. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  59. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  60. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Pipeline parallelism: Run next merging task as soon as possible Core 1 Core 2 Core 3 Core 4 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  61. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Pipeline parallelism: Run next merging task as soon as possible Core 1 Core 2 Core 3 Core 4 Even more speedup Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  62. Pipelined parallel mergesort Classic parallelism: start a task when the previous one is done Core 1 Core 2 Core 3 Core 4 Pipeline parallelism: Run next merging task as soon as possible Core 1 Core 2 Core 3 Core 4 Even more speedup Difficult to implement manually Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 29 / 40

  63. Pipeline parallelism Related research since the 60’ Program verifiability Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 30 / 40

  64. Pipeline parallelism Related research since the 60’ Program verifiability Parallelism is a mere “consequence” Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 30 / 40

  65. Pipeline parallelism Related research since the 60’ Program verifiability Parallelism is a mere “consequence” Sequential tasks communicating through channels Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 30 / 40

  66. Pipeline parallelism Related research since the 60’ Program verifiability Parallelism is a mere “consequence” Sequential tasks communicating through channels Theories: Kahn Networks, (Synchronous) Data Flow, Communicating Sequential Processes 1 1 1 1 + + 1 1 1 1 D 1 1 1 1 *k *k D *k *k 1 1 1 1 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 30 / 40

  67. Pipeline parallelism Related research since the 60’ Program verifiability Parallelism is a mere “consequence” Sequential tasks communicating through channels Theories: Kahn Networks, (Synchronous) Data Flow, Communicating Sequential Processes Languages: Streamit, CAL, Esterel 1 1 1 1 + + 1 1 1 1 D 1 1 1 1 *k *k D *k *k 1 1 1 1 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 30 / 40

  68. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  69. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  70. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  71. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Stream programming (Mostly) sequential tasks Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  72. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Stream programming (Mostly) sequential tasks Actual parallelism: scheduling Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  73. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Stream programming (Mostly) sequential tasks Actual parallelism: scheduling No universal shared memory Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  74. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Stream programming (Mostly) sequential tasks Actual parallelism: scheduling No universal shared memory Natural to pipeline parallelism Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  75. Classic versus stream Most programming languages unsuitable to parallelism Abstract a single, universal instruction pointer Abstract a single, universal address space Difficult to read with several threads in mind Annotations (OpenMP) not helping with high number of cores Stream programming (Mostly) sequential tasks Actual parallelism: scheduling No universal shared memory Natural to pipeline parallelism Communications with on-chip memories: on-chip pipelining Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 31 / 40

  76. Back to parallel merge Classic parallelism Core 1 Core 2 Core 3 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 32 / 40

  77. Back to parallel merge Classic parallelism Core 1 Core 2 Core 3 Pipelining (4 initial sorting tasks) Core 1 Core 2 Core 3 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 32 / 40

  78. Back to parallel merge Classic parallelism Core 1 Core 2 Core 3 Pipelining (4 initial sorting tasks) Core 1 Core 2 Core 3 Pipelining (8 initial sorting tasks ) Core 1 Core 2 Core 3 Nicolas Melot nicolas.melot (at) liu.se (LIU) TDDD56 lesson 1 November 4, 2015 32 / 40

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Multicore DSP Architecture and Programming O. Dahl 1 1 Electrical Engineering, Linkping

Multicore DSP Architecture and Programming O. Dahl 1 1 Electrical Engineering, Linkping

System requirements System design System development Summary Multicore DSP Architecture and Programming O. Dahl 1 1 Electrical Engineering, Linkping University, Linkping, Sweden Guest lecture in TDDD56 Multicore and GPU Programming, LiU,

626 views • 58 slides
TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se Labs

TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se Labs

TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se Labs schedule WebReg Week Lab 1 v46 Load Balancing Responsible: August Lab 2 v47 CPU Non-blocking Data Structures Lesson 2 Lab 3 v48 High-level

657 views • 21 slides
CS 240A: Shared Memory &amp; Multicore Programming with Cilk++ Multicore and NUMA

CS 240A: Shared Memory &amp; Multicore Programming with Cilk++ Multicore and NUMA

CS 240A: Shared Memory &amp; Multicore Programming with Cilk++ Multicore and NUMA architectures Multithreaded Programming Cilk++ as a concurrency platform Work, Span, (potential) Parallelism Thanks to to Charles E. E. Leiserson

849 views • 45 slides
TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se C++11

TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se C++11

TDDD56 Lab 3: Skeleton programming with SkePU August Ernstsson august.ernstsson@liu.se C++11 Shift in the labs from C to C++11 (modern C++) // auto type specifier auto addOneMap = skepu2::Map&lt;1&gt;(addOneFunc);

1.06k views • 19 slides
Theory Exercises Exercise 1 Consider a simple uniprocessor system with no caches. How does

Theory Exercises Exercise 1 Consider a simple uniprocessor system with no caches. How does

TDDD56 Multicore and GPU Programming 2013 Theory Exercises Exercise 1 Consider a simple uniprocessor system with no caches. How does register allocation (applied by the compiler) affect memory consistency? Which language feature of C allows you

429 views • 3 slides
T-106.5800 Seminar on Software Techniques Seminar on Multicore Programming Multicore Technology

T-106.5800 Seminar on Software Techniques Seminar on Multicore Programming Multicore Technology

T-106.5800 Seminar on Software Techniques Seminar on Multicore Programming Multicore Technology in Mobile Devices Antti P Miettinen antti.p.miettinen@nokia.com February 12, 2009 Abstract power budget of roughly three Watts for a hand- held

741 views • 4 slides
Multicore Semantics and Programming Tim Harris Peter Sewell Amazon University of Cambridge

Multicore Semantics and Programming Tim Harris Peter Sewell Amazon University of Cambridge

Multicore Semantics and Programming Tim Harris Peter Sewell Amazon University of Cambridge October November, 2019 These Lectures Part 1: Multicore Programming: Concurrent algorithms (Tim Harris, Amazon) Concurrent programming: simple

4.69k views • 166 slides
Multicore Programming: C++0x Mark Batty University of Cambridge in collaboration with Scott

Multicore Programming: C++0x Mark Batty University of Cambridge in collaboration with Scott

Multicore Programming: C++0x Mark Batty University of Cambridge in collaboration with Scott Owens, Susmit Sarkar, Peter Sewell, Tjark Weber November, 2010 p. 1 C++0x: the next C++ Specified by the C++ Standards Committee Defined in The

550 views • 38 slides
Programming a multicore architecture without coherency and atomic operations Jochem Rutgers ,

Programming a multicore architecture without coherency and atomic operations Jochem Rutgers ,

Programming a multicore architecture without coherency and atomic operations Jochem Rutgers , Marco Bekooij, Gerard Smit 2014-02-15 1 / 13 Parallel render example One master thread: data = read_3d_model_from_file(); 1 go = 1; 2 while

693 views • 34 slides
Introduction to Multicore Programming Marc Moreno Maza University of Western Ontario, London,

Introduction to Multicore Programming Marc Moreno Maza University of Western Ontario, London,

Introduction to Multicore Programming Marc Moreno Maza University of Western Ontario, London, Ontario (Canada) CS 4402 - CS 9535 (Moreno Maza) Introduction to Multicore Programming CS 4402 - CS 9535 1 / 60 Plan Multi-core Architecture 1

358 views • 5 slides
System on Chip C (SoC-C) Efficient programming abstractions for heterogeneous multicore Systems

System on Chip C (SoC-C) Efficient programming abstractions for heterogeneous multicore Systems

System on Chip C (SoC-C) Efficient programming abstractions for heterogeneous multicore Systems on Chip Alastair Reid ARM Ltd Yuan Lin University of Michigan Krisztian Flautner ARM Ltd Edmund Grimley-Evans ARM Ltd 1 Mobile Consumer

679 views • 52 slides
Programming Tools for Embedded Multicore Jakob Engblom Technical Marketing Manager Simics

Programming Tools for Embedded Multicore Jakob Engblom Technical Marketing Manager Simics

Programming Tools for Embedded Multicore Jakob Engblom Technical Marketing Manager Simics Wind River jakob.engblom@windriver.com | http://blogs.windriver.com/engblom/ Disclaimer These are my personal views on multicore and embedded

514 views • 33 slides
Multicore Programming Java Memory Model Jaroslav ev Peter Sewell ck Tim Harris

Multicore Programming Java Memory Model Jaroslav ev Peter Sewell ck Tim Harris

Multicore Programming Java Memory Model Jaroslav ev Peter Sewell ck Tim Harris University of Cambridge MSR with thanks to Francesco Zappa Nardelli, Susmit Sarkar, Tom Ridge, Scott Owens, Magnus O. Myreen, Luc Maranget, Mark Batty,

901 views • 74 slides
The Why, Where and How of Multicore Anant Agarwal MIT and Tilera Corp. What is Multicore?

The Why, Where and How of Multicore Anant Agarwal MIT and Tilera Corp. What is Multicore?

The Why, Where and How of Multicore Anant Agarwal MIT and Tilera Corp. What is Multicore? Whatevers Inside? What is Multicore? Whatevers Inside? Seriously, multicore satisfies three properties Single chip Multiple

494 views • 35 slides
PROGRAMMING Programming Mission Statement Lesson Planning Welcome! Club Mission, Awesome

PROGRAMMING Programming Mission Statement Lesson Planning Welcome! Club Mission, Awesome

PROGRAMMING Programming Mission Statement Lesson Planning Welcome! Club Mission, Awesome Introductions What are we doing today? Read a Book about it GO GET DIRTY/Hands-on experience Harvest Snack creation time! What

311 views • 30 slides
Embedded System Programming Multicore ES (Module 40) Yann-Hang Lee Arizona State University

Embedded System Programming Multicore ES (Module 40) Yann-Hang Lee Arizona State University

Embedded System Programming Multicore ES (Module 40) Yann-Hang Lee Arizona State University yhlee@asu.edu (480) 727-7507 Summer 2014 Real-time Systems Lab, Computer Science and Engineering, ASU The Era of Multi-core Processors thread

436 views • 12 slides
Xen Summit 18 April, 2007 TJ Watson Research Center The Xen-API Ewan Mellor ewan@xensource.com

Xen Summit 18 April, 2007 TJ Watson Research Center The Xen-API Ewan Mellor ewan@xensource.com

Xen Summit 18 April, 2007 TJ Watson Research Center The Xen-API Ewan Mellor ewan@xensource.com Xen Management Architecture Cluster-wide Integration The aim is to integrate Xen-based systems with enterprise-level cluster management

213 views • 17 slides
Improving Model Selection by Employing the Test Data Max Westphal, Werner Brannath University of

Improving Model Selection by Employing the Test Data Max Westphal, Werner Brannath University of

3 Improving Model Selection by Employing the Test Data Max Westphal, Werner Brannath University of Bremen, Germany Institute for Statistics Research Training Group 3 mwestphal@uni-bremen.de https://github.com/maxwestphal/ ICML 2019, Long

430 views • 13 slides
The evolutions of spinning bodies moving in rotating black hole spacetimes Zoltn Keresztes

The evolutions of spinning bodies moving in rotating black hole spacetimes Zoltn Keresztes

The evolutions of spinning bodies moving in rotating black hole spacetimes Zoltn Keresztes Balzs Mikczi Department of Theoretical Physics, University of Szeged Research Institute for Particle and Nuclear Physics, Wigner RCP The work

527 views • 20 slides
Practical Enterprise Integration Realising the Benefits of a Strong Canonical Architecture Andrew

Practical Enterprise Integration Realising the Benefits of a Strong Canonical Architecture Andrew

Practical Enterprise Integration Realising the Benefits of a Strong Canonical Architecture Andrew K. Johnston Independent consultant at National Grid www.andrewj.com www.agilearchitect.com &lt;Contributors&gt; www.nationalgrid.com 2 EAC11

805 views • 23 slides
Fourier-pseudospectral method for Cahn-Hilliard Equation on GPU Kangping Zhu Courant Institute

Fourier-pseudospectral method for Cahn-Hilliard Equation on GPU Kangping Zhu Courant Institute

Fourier-pseudospectral method for Cahn-Hilliard Equation on GPU Kangping Zhu Courant Institute of Mathematical Sciences kangping@cims.nyu.edu December 18, 2012 Kangping Zhu (CIMS) CH equation on GPU December 18, 2012 1 / 16 Background

398 views • 16 slides
Reduced Manifolds and Trajectory Curvature J. M. Powers Department of Aerospace and Mechanical

Reduced Manifolds and Trajectory Curvature J. M. Powers Department of Aerospace and Mechanical

Reduced Manifolds and Trajectory Curvature J. M. Powers Department of Aerospace and Mechanical Engineering University of Notre Dame, USA Sixth International Workshop on Model Reduction in Reacting Flows Princeton, New Jersey 12 July 2017 6th

327 views • 19 slides
Longstanding Discrepancies in Stratospheric Water Vapor Measurements Revisited During the 2011

Longstanding Discrepancies in Stratospheric Water Vapor Measurements Revisited During the 2011

Longstanding Discrepancies in Stratospheric Water Vapor Measurements Revisited During the 2011 Mid-latitude Airborne Cirrus Properties Experiment (MACPEX) Dale Hurst Jessica Smith Emrys Hall Maryann Sargent Allen Jordan David Sayres Troy

311 views • 14 slides
How can different campuses inter-operate? From ora et labora to collaborate and federate

How can different campuses inter-operate? From ora et labora to collaborate and federate

How can different campuses inter-operate? From ora et labora to collaborate and federate EuroCAMP, Ljubljana, 2006-04-03 Ingrid Melve, FEIDE manager From S S O to a federated solution Campus Identity Management Collaborations

678 views • 33 slides

More recommend