constantin
/
bachelor-thesis
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
This contains my bachelors thesis and associated tex files, code snippets and maybe more. Topic: Data Movement in Heterogeneous Memories with Intel Data Streaming Accelerator
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
538 Commits
1 Branch
0 Tags
92 MiB
Branch: master
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'master'
${ noResults }
bachelor-thesis/benchmarks/benchmark-descriptors/peak-perf-8dsa/copy-n0ton11-1gib-8dsa.json

102 lines
1.7 KiB
Raw Permalink Normal View History

move the existing results into a new folder and modify them to be only benchmark descriptors
1 year ago
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
use a batch size of 8 to check whether multiple engines can increase throughput
1 year ago
​
​
​
​
fix an issue in the python script that lead to references being modified which caused bad node settings
1 year ago
prepare peak throughput plotter for multi-node results
1 year ago
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
prepare peak throughput plotter for multi-node results
1 year ago
​
​
​
​
​
fix an issue in the python script that lead to references being modified which caused bad node settings
1 year ago
prepare peak throughput plotter for multi-node results
1 year ago
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
prepare peak throughput plotter for multi-node results
1 year ago
​
​
​
​
​
fix an issue in the python script that lead to references being modified which caused bad node settings
1 year ago
prepare peak throughput plotter for multi-node results
1 year ago
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
prepare peak throughput plotter for multi-node results
1 year ago
​
​
​
​
​
use different engine configuration depending on whether intra socket (all 4 engines on the socket) or inter socket (src and destination engine, cross copy) is the copy type
1 year ago
use a batch size of 8 to check whether multiple engines can increase throughput
1 year ago
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
use a batch size of 8 to check whether multiple engines can increase throughput
1 year ago
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
​
​
​
​
​
​
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
​
​
​
​
​
​
​
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
​
​
​
​
​
​
​
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
​
​
​
​
​
​
​
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
add option for internal repetitions to benchmarks which allows the small copies of 1kib to run long enough for the timings to become usable (goal is about 1s runtime for each iteration)
11 months ago
​
provide two benchmarks for peak performance, one that is brute force and one that uses smart node assignment and therefore lower utilization
1 year ago
use a batch size of 8 to check whether multiple engines can increase throughput
1 year ago
​
rewrite descriptors to match the format of the rewrite of bench from previous commit
11 months ago
​
move the existing results into a new folder and modify them to be only benchmark descriptors
1 year ago
  1. {
  2. "count": 8,
  3. "list": [
  4. {
  5. "affinity": {
  6. "nnode_dst": 11,
  7. "nnode_src": 0,
  8. "node": 0
  9. },
  10. "task": {
  11. "size": 1073741824,
  12. "batch_size": 0,
  13. "reps": 1
  14. }
  15. },
  16. {
  17. "affinity": {
  18. "nnode_dst": 11,
  19. "nnode_src": 0,
  20. "node": 1
  21. },
  22. "task": {
  23. "size": 1073741824,
  24. "batch_size": 0,
  25. "reps": 1
  26. }
  27. },
  28. {
  29. "affinity": {
  30. "nnode_dst": 11,
  31. "nnode_src": 0,
  32. "node": 2
  33. },
  34. "task": {
  35. "size": 1073741824,
  36. "batch_size": 0,
  37. "reps": 1
  38. }
  39. },
  40. {
  41. "affinity": {
  42. "nnode_dst": 11,
  43. "nnode_src": 0,
  44. "node": 3
  45. },
  46. "task": {
  47. "size": 1073741824,
  48. "batch_size": 0,
  49. "reps": 1
  50. }
  51. },
  52. {
  53. "affinity": {
  54. "nnode_dst": 11,
  55. "nnode_src": 0,
  56. "node": 4
  57. },
  58. "task": {
  59. "size": 1073741824,
  60. "batch_size": 0,
  61. "reps": 1
  62. }
  63. },
  64. {
  65. "affinity": {
  66. "nnode_dst": 11,
  67. "nnode_src": 0,
  68. "node": 5
  69. },
  70. "task": {
  71. "size": 1073741824,
  72. "batch_size": 0,
  73. "reps": 1
  74. }
  75. },
  76. {
  77. "affinity": {
  78. "nnode_dst": 11,
  79. "nnode_src": 0,
  80. "node": 6
  81. },
  82. "task": {
  83. "size": 1073741824,
  84. "batch_size": 0,
  85. "reps": 1
  86. }
  87. },
  88. {
  89. "affinity": {
  90. "nnode_dst": 11,
  91. "nnode_src": 0,
  92. "node": 7
  93. },
  94. "task": {
  95. "size": 1073741824,
  96. "batch_size": 0,
  97. "reps": 1
  98. }
  99. }
  100. ],
  101. "path": "hw",
  102. "repetitions": 10
  103. }