parallel_programming_course/tasks/seq/broken_example_disabled/perf_tests/main.cpp at 539669a3aaf9efbe8e00c33709268f38c6f1c3f0 · learning-process/parallel_programming_course · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
#include <gtest/gtest.h>

#include <chrono>
#include <cstddef>
#include <cstdint>
#include <memory>
#include <vector>

#include "core/perf/include/perf.hpp"
#include "core/task/include/task.hpp"
#include "seq/example/include/ops_seq.hpp"

TEST(nesterov_a_test_task_seq, test_pipeline_run) {
  constexpr int kCount = 500;

  // Create data
  std::vector<int> in(kCount * kCount, 0);
  std::vector<int> out(kCount * kCount, 0);

  for (size_t i = 0; i < kCount; i++) {
    in[(i * kCount) + i] = 1;
  }

  // Create task_data
  auto task_data_seq = std::make_shared<ppc::core::TaskData>();
  task_data_seq->inputs.emplace_back(reinterpret_cast<uint8_t *>(in.data()));
  task_data_seq->inputs_count.emplace_back(in.size());
  task_data_seq->outputs.emplace_back(reinterpret_cast<uint8_t *>(out.data()));
  task_data_seq->outputs_count.emplace_back(out.size());

  // Create Task
  auto test_task_sequential = std::make_shared<nesterov_a_test_task_seq::TestTaskSequential>(task_data_seq);

  // Create Perf attributes
  auto perf_attr = std::make_shared<ppc::core::PerfAttr>();
  perf_attr->num_running = 10;
  const auto t0 = std::chrono::high_resolution_clock::now();
  perf_attr->current_timer = [&] {
    auto current_time_point = std::chrono::high_resolution_clock::now();
    auto duration = std::chrono::duration_cast<std::chrono::nanoseconds>(current_time_point - t0).count();
    return static_cast<double>(duration) * 1e-9;
  };

  // Create and init perf results
  auto perf_results = std::make_shared<ppc::core::PerfResults>();

  // Create Perf analyzer
  auto perf_analyzer = std::make_shared<ppc::core::Perf>(test_task_sequential);
  perf_analyzer->PipelineRun(perf_attr, perf_results);
  ppc::core::Perf::PrintPerfStatistic(perf_results);
  ASSERT_EQ(in, out);
}

TEST(nesterov_a_test_task_seq, test_task_run) {
  constexpr int kCount = 500;

  // Create data
  std::vector<int> in(kCount * kCount, 0);
  std::vector<int> out(kCount * kCount, 0);

  for (size_t i = 0; i < kCount; i++) {
    in[(i * kCount) + i] = 1;
  }

  // Create task_data
  auto task_data_seq = std::make_shared<ppc::core::TaskData>();
  task_data_seq->inputs.emplace_back(reinterpret_cast<uint8_t *>(in.data()));
  task_data_seq->inputs_count.emplace_back(in.size());
  task_data_seq->outputs.emplace_back(reinterpret_cast<uint8_t *>(out.data()));
  task_data_seq->outputs_count.emplace_back(out.size());

  // Create Task
  auto test_task_sequential = std::make_shared<nesterov_a_test_task_seq::TestTaskSequential>(task_data_seq);

  // Create Perf attributes
  auto perf_attr = std::make_shared<ppc::core::PerfAttr>();
  perf_attr->num_running = 10;
  const auto t0 = std::chrono::high_resolution_clock::now();
  perf_attr->current_timer = [&] {
    auto current_time_point = std::chrono::high_resolution_clock::now();
    auto duration = std::chrono::duration_cast<std::chrono::nanoseconds>(current_time_point - t0).count();
    return static_cast<double>(duration) * 1e-9;
  };

  // Create and init perf results
  auto perf_results = std::make_shared<ppc::core::PerfResults>();

  // Create Perf analyzer
  auto perf_analyzer = std::make_shared<ppc::core::Perf>(test_task_sequential);
  perf_analyzer->TaskRun(perf_attr, perf_results);
  ppc::core::Perf::PrintPerfStatistic(perf_results);
  ASSERT_EQ(in, out);
}