Files
conflict-set/RealDataBench.cpp

148 lines
4.3 KiB
C++

#include <ConflictSet.h>
#include <cerrno>
#include <chrono>
#include <cstdio>
#include <cstring>
#include <fcntl.h>
#include <span>
#include <sys/mman.h>
#include <sys/stat.h>
#include <unistd.h>
#include <vector>
using namespace weaselab;
double now() {
return std::chrono::duration_cast<std::chrono::nanoseconds>(
std::chrono::steady_clock::now().time_since_epoch())
.count() *
1e-9;
}
inline size_t getPageSize() {
static size_t kPageSize = sysconf(_SC_PAGESIZE);
return kPageSize;
}
/// Helper for rounding up to page size (or some other alignment)
constexpr inline size_t rightAlign(size_t offset, size_t alignment) {
return offset % alignment == 0 ? offset
: ((offset / alignment) + 1) * alignment;
}
int main(int argc, const char **argv) {
// Use with this dataset https://snap.stanford.edu/data/memetracker9.html
// Preprocess the files with `sed -i'' '/^Q/d'`
double checkTime = 0;
double addTime = 0;
double gcTime = 0;
double checkBytes = 0;
double addBytes = 0;
ConflictSet cs{0};
int64_t version = 0;
double timer = 0;
int64_t peakMemory = 0;
for (int i = 1; i < argc; ++i) {
int fd = open(argv[i], O_RDONLY);
struct stat st;
if (fstat(fd, &st) == -1) {
int err = errno;
fprintf(stderr, "stat error %s - %s\n", argv[i], strerror(err));
fflush(stderr);
abort();
}
int64_t size = rightAlign(st.st_size, getPageSize());
const uint8_t *begin =
(uint8_t *)mmap(0, size, PROT_READ, MAP_PRIVATE, fd, 0);
madvise((void *)begin, size, MADV_SEQUENTIAL);
auto *const mapOriginal = begin;
const auto sizeOriginal = size;
using StringView = std::span<const uint8_t>;
StringView write;
std::vector<StringView> reads;
std::vector<ConflictSet::ReadRange> readRanges;
std::vector<ConflictSet::Result> results;
for (uint8_t *end = (uint8_t *)memchr(begin, '\n', size); end != nullptr;) {
StringView line{begin, static_cast<size_t>(end - begin)};
size -= end - begin + 1;
begin = end + 1;
end = (uint8_t *)memchr(begin, '\n', size);
if (line.size() > 0 && line[0] == 'P') {
write = line.subspan(2, line.size());
} else if (line.size() > 0 && line[0] == 'L') {
reads.push_back(line.subspan(2, line.size()));
} else if (line.empty()) {
{
readRanges.resize(reads.size());
auto iter = readRanges.begin();
for (const auto &read : reads) {
iter->begin.p = (const uint8_t *)read.data();
iter->begin.len = read.size();
checkBytes += read.size();
iter->end.len = 0;
iter->readVersion = version - 100;
++iter;
}
}
results.resize(readRanges.size());
timer = now();
cs.check(readRanges.data(), results.data(), readRanges.size());
checkTime += now() - timer;
// Add unconditionally so that the load doesn't actually depend on the
// conflict rate
ConflictSet::WriteRange w;
w.begin.p = (const uint8_t *)write.data();
w.begin.len = write.size();
w.end.len = 0;
addBytes += write.size();
timer = now();
cs.addWrites(&w, 1, ++version);
addTime += now() - timer;
write = {};
reads.clear();
if (cs.getBytes() > peakMemory) {
peakMemory = cs.getBytes();
}
timer = now();
cs.setOldestVersion(version - 10000);
gcTime += now() - timer;
}
}
munmap((void *)mapOriginal, sizeOriginal);
close(fd);
}
ConflictSet::MetricsV1 *metrics;
int metricsCount;
cs.getMetricsV1(&metrics, &metricsCount);
for (int i = 0; i < metricsCount; ++i) {
fprintf(stderr, "# HELP %s %s\n", metrics[i].name, metrics[i].help);
fprintf(stderr, "# TYPE %s %s\n", metrics[i].name,
metrics[i].type == metrics[i].Counter ? "counter" : "gauge");
fprintf(stderr, "%s %g\n", metrics[i].name, metrics[i].getValue());
}
printf("Check: %g seconds, %g MB/s, Add: %g seconds, %g MB/s, Gc ratio: "
"%g%%, Peak idle memory: %g\n",
checkTime, checkBytes / checkTime * 1e-6, addTime,
addBytes / addTime * 1e-6, gcTime / (gcTime + addTime) * 1e2,
double(peakMemory));
}