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
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
|
#include "afio_pch.hpp"
#ifdef _DEBUG
#define ITEMS 64
#else
#define ITEMS 65536
#endif
#define PARALLELISM 256 // up to max fds on your platform
namespace iostreams {
#include "workshop_naive.ipp"
}
namespace naive {
#include "workshop_naive_afio.ipp"
}
namespace atomic_updates {
#include "workshop_atomic_updates_afio.ipp"
}
namespace final {
#include "workshop_final_afio.ipp"
#include "../detail/SpookyV2.cpp"
}
namespace filesystem = BOOST_AFIO_V2_NAMESPACE::filesystem;
namespace chrono = BOOST_AFIO_V2_NAMESPACE::chrono;
using BOOST_AFIO_V2_NAMESPACE::ratio;
// From http://burtleburtle.net/bob/rand/smallprng.html
typedef unsigned int u4;
typedef struct ranctx { u4 a; u4 b; u4 c; u4 d; } ranctx;
#define rot(x,k) (((x)<<(k))|((x)>>(32-(k))))
static u4 ranval(ranctx *x) {
u4 e = x->a - rot(x->b, 27);
x->a = x->b ^ rot(x->c, 17);
x->b = x->c + x->d;
x->c = x->d + e;
x->d = e + x->a;
return x->d;
}
static void raninit(ranctx *x, u4 seed) {
u4 i;
x->a = 0xf1ea5eed, x->b = x->c = x->d = seed;
for (i = 0; i < 20; ++i) {
(void) ranval(x);
}
}
static std::vector<std::pair<ranctx, unsigned>> items;
void prepare()
{
ranctx gen;
raninit(&gen, 0x78adbcff);
items.clear();
items.reserve(ITEMS);
for (size_t n = 0; n < ITEMS; n++)
{
unsigned t=ranval(&gen);
items.push_back(std::make_pair(gen, t));
}
}
template<class data_store> void benchmark(const char *filename, const char *desc, bool parallel_writes)
{
typedef chrono::duration<double, ratio<1, 1>> secs_type;
std::vector<std::tuple<size_t, double, double>> results;
prepare();
for(size_t n=1; n<=ITEMS; n<<=1)
{
std::cout << "Benchmarking " << desc << " insertion of " << n << " records ... ";
if(filesystem::exists("store"))
filesystem::remove_all("store");
auto begin=chrono::high_resolution_clock::now();
while(chrono::duration_cast<secs_type>(chrono::high_resolution_clock::now()-begin).count()<1);
data_store store(data_store::writeable);
size_t write_parallelism = parallel_writes ? PARALLELISM : 1;
std::vector<typename data_store::write_result_type> ops(write_parallelism);
begin=chrono::high_resolution_clock::now();
for (size_t m = 0; m < n; m += write_parallelism)
{
int todo = (int)(n < write_parallelism ? n : write_parallelism);
#pragma omp parallel for
for (int o = 0; o < todo; o++)
ops[o] = store.write(std::to_string(m + (size_t)o));
#pragma omp parallel for
for (int o = 0; o < todo; o++)
{
*ops[o].get() << items[m + (size_t)o].second;
ops[o].get()->flush();
}
// Some dumping of the writes should have happened by now
#pragma omp parallel for
for (int o = 0; o < todo; o++)
ops[o]=typename data_store::write_result_type();
}
auto end=chrono::high_resolution_clock::now();
auto diff=chrono::duration_cast<secs_type>(end-begin).count();
std::cout << (n/diff) << " items/sec" << std::endl;
results.push_back(std::make_tuple(n, n/diff, 0));
}
auto resultsit = results.begin();
for (size_t n = 1; n <= ITEMS; n <<= 1)
{
std::cout << "Benchmarking " << desc << " lookup of " << n << " records ... ";
data_store store;
std::vector<typename data_store::lookup_result_type> ops(PARALLELISM);
auto begin = chrono::high_resolution_clock::now();
for (size_t m = 0; m < n; m += PARALLELISM)
{
int todo = n < PARALLELISM ? n : PARALLELISM;
#pragma omp parallel for
for (int o = 0; o < todo; o++)
ops[o] = store.lookup(std::to_string(m + (size_t) o));
// Some readahead should have happened by now
#pragma omp parallel for
for (int o = 0; o < todo; o++)
{
auto s(ops[o].get());
unsigned t;
*s >> t;
if (t != items[m+(size_t) o].second)
std::cerr << "ERROR: Item " << m << " has incorrect contents!" << std::endl;
ops[o]=typename data_store::lookup_result_type();
}
}
auto end = chrono::high_resolution_clock::now();
auto diff = chrono::duration_cast<secs_type>(end - begin).count();
std::cout << (n / diff) << " items/sec" << std::endl;
std::get<2>(*resultsit++) = n / diff;
}
std::ofstream csvh(filename);
csvh << "Items,Inserts/sec,Lookups/sec" << std::endl;
for(auto &i : results)
csvh << std::get<0>(i) << "," << std::get<1>(i) << "," << std::get<2>(i) << std::endl;
}
int main(void)
{
typedef chrono::duration<double, ratio<1, 1>> secs_type;
auto begin=chrono::high_resolution_clock::now();
while(chrono::duration_cast<secs_type>(chrono::high_resolution_clock::now()-begin).count()<3);
//benchmark<iostreams::data_store>("iostreams.csv", "STL iostreams", true);
//benchmark<naive::data_store>("afio_naive.csv", "AFIO naive", true);
//benchmark<atomic_updates::data_store>("afio_atomic.csv", "AFIO atomic update", true);
benchmark<final::data_store>("afio_final.csv", "AFIO single file", true);
return 0;
}
|
