Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Reduce Histogram Cardinality #595

Merged
merged 7 commits into from
Jan 8, 2023
Merged

Reduce Histogram Cardinality #595

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
a2098f6
Reduce Histogram Cardinality
Jan 6, 2023
97e22d3
use std::unique
Jan 6, 2023
cab8165
Debug macos
Jan 6, 2023
93a4888
fix debug build
Jan 6, 2023
52c9b42
Test zero initialize
Jan 6, 2023
f20dbe6
change type
Jan 6, 2023
e711050
Remove debug print
Jan 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
86 changes: 52 additions & 34 deletions src/Metrics.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,11 +21,17 @@
#pragma GCC diagnostic pop
#endif
#include <chrono>
#include <math.h>
#include <regex>
#include <set>
#include <shared_mutex>
#include <vector>

#define HIST_MIN_EXP -9
#define HIST_MAX_EXP 18
#define HIST_LOG_BUCK 18
#define HIST_N_BUCKETS (HIST_LOG_BUCK * (HIST_MAX_EXP - HIST_MIN_EXP))

namespace visor {

using json = nlohmann::json;
Expand Down Expand Up @@ -176,18 +182,36 @@ template <typename T>
class Histogram final : public Metric
{
static_assert(std::is_integral<T>::value || std::is_floating_point<T>::value);
T _pace;

static constexpr T _get_pace()
{
if constexpr (std::is_integral<T>::value) {
return 1;
} else {
return 0.0000001;
}
}

// calculated at compile time
static constexpr std::pair<std::array<T, HIST_N_BUCKETS>, size_t> _get_boundaries()
{
auto pace = _get_pace();
std::array<T, HIST_N_BUCKETS> boundaries{};
size_t index = 0;
for (auto exponent = HIST_MIN_EXP; exponent < HIST_MAX_EXP; exponent++) {
for (auto buckets = 0; buckets < HIST_LOG_BUCK; buckets++) {
boundaries[index++] = static_cast<T>((std::pow(10.0, static_cast<float>(buckets) / HIST_LOG_BUCK) * std::pow(10.0, exponent)) + pace);
}
}
auto itr = std::unique(boundaries.begin(), boundaries.end());
return {boundaries, std::distance(boundaries.begin(), itr)};
}
datasketches::kll_sketch<T> _sketch;

public:
Histogram(std::string schema_key, std::initializer_list<std::string> names, std::string desc)
: Metric(schema_key, names, std::move(desc))
{
if constexpr (std::is_integral<T>::value) {
_pace = 1;
} else {
_pace = 0.0000001;
}
}

void update(const T &value)
Expand Down Expand Up @@ -226,52 +250,47 @@ class Histogram final : public Metric
if (_sketch.is_empty()) {
return;
}
std::size_t split_point_size = 1 + static_cast<std::size_t>(std::log2(_sketch.get_n())); // Sturge’s Rule
T step = _sketch.get_max_value() / split_point_size;
T split_point = step;
std::unique_ptr<T[]> bins = std::make_unique<T[]>(split_point_size);
for (std::size_t i = 0; i < split_point_size; ++i) {
bins[i] = split_point + _pace;
split_point += step;
if (split_point > _sketch.get_max_value()) {
split_point = _sketch.get_max_value();
auto bins_pmf = _get_boundaries();
auto histogram_pmf = _sketch.get_PMF(bins_pmf.first.data(), bins_pmf.second);
std::vector<T> bins;
for (size_t i = 0; i < bins_pmf.second; ++i) {
if (histogram_pmf[i]) {
bins.push_back(bins_pmf.first[i]);
}
}
auto histogram = _sketch.get_CDF(bins.get(), split_point_size);
for (std::size_t i = 0; i < split_point_size; ++i) {
name_json_assign(j, {"buckets", std::to_string(bins[i] - _pace)}, histogram[i] * _sketch.get_n());
auto histogram = _sketch.get_CDF(bins.data(), bins.size());
auto pace = _get_pace();
for (std::size_t i = 0; i < bins.size(); ++i) {
name_json_assign(j, {"buckets", std::to_string(bins[i] - pace)}, histogram[i] * _sketch.get_n());
}
name_json_assign(j, {"buckets", "+Inf"}, histogram[split_point_size] * _sketch.get_n());
name_json_assign(j, {"buckets", "+Inf"}, histogram[bins.size()] * _sketch.get_n());
}

void to_prometheus(std::stringstream &out, Metric::LabelMap add_labels = {}) const override
{
if (_sketch.is_empty()) {
return;
}
std::size_t split_point_size = 1 + static_cast<std::size_t>(std::log2(_sketch.get_n())); // Sturge’s Rule
T step = _sketch.get_max_value() / split_point_size;
T split_point = step;
std::unique_ptr<T[]> bins = std::make_unique<T[]>(split_point_size);
for (std::size_t i = 0; i < split_point_size; ++i) {
bins[i] = split_point + _pace;
split_point += step;
if (split_point > _sketch.get_max_value()) {
split_point = _sketch.get_max_value();
auto bins_pmf = _get_boundaries();
auto histogram_pmf = _sketch.get_PMF(bins_pmf.first.data(), bins_pmf.second);
std::vector<T> bins;
for (size_t i = 0; i < bins_pmf.second; ++i) {
if (histogram_pmf[i]) {
bins.push_back(bins_pmf.first[i]);
}
}
auto histogram = _sketch.get_CDF(bins.get(), split_point_size);

auto histogram = _sketch.get_CDF(bins.data(), bins.size());
auto pace = _get_pace();
out << "# HELP " << base_name_snake() << ' ' << _desc << std::endl;
out << "# TYPE " << base_name_snake() << " histogram" << std::endl;
for (std::size_t i = 0; i < split_point_size; ++i) {
for (std::size_t i = 0; i < bins.size(); ++i) {
LabelMap le(add_labels);
le["le"] = std::to_string(bins[i] - _pace);
le["le"] = std::to_string(bins[i] - pace);
out << name_snake({"bucket"}, le) << ' ' << histogram[i] * _sketch.get_n() << std::endl;
}
LabelMap le(add_labels);
le["le"] = "+Inf";
out << name_snake({"bucket"}, le) << ' ' << histogram[split_point_size] * _sketch.get_n() << std::endl;
out << name_snake({"bucket"}, le) << ' ' << histogram[bins.size()] * _sketch.get_n() << std::endl;
out << name_snake({"count"}, add_labels) << ' ' << _sketch.get_n() << std::endl;
}

Expand Down Expand Up @@ -872,5 +891,4 @@ class Rate final : public Metric
void to_prometheus(std::stringstream &out, Metric::LabelMap add_labels = {}) const override;
void to_opentelemetry(metrics::v1::ScopeMetrics &scope, timespec &start, timespec &end, LabelMap add_labels = {}) const override;
};

}
19 changes: 8 additions & 11 deletions src/tests/test_metrics.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -254,13 +254,13 @@ TEST_CASE("Histogram int metrics", "[metrics][histogram]")
std::stringstream output;
metrics::v1::ScopeMetrics scope;
std::string line;
Histogram<int_fast32_t> h("root", {"test", "metric"}, "A histogram test metric");
Histogram<uint64_t> h("root", {"test", "metric"}, "A histogram test metric");

SECTION("Histogram to json")
{
h.name_json_assign(j, 58);
CHECK(j["test"]["metric"] == 58);
int_fast32_t value = 12;
uint64_t value = 12;
h.update(value);
h.update(value);
h.update(value);
Expand All @@ -270,7 +270,6 @@ TEST_CASE("Histogram int metrics", "[metrics][histogram]")

CHECK(j["top"]["test"]["metric"]["buckets"]["+Inf"] == 4.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["12"] == 4.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["4"] == 0.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["8"] == 1.0);
}

Expand All @@ -294,7 +293,7 @@ TEST_CASE("Histogram int metrics", "[metrics][histogram]")
std::getline(output, line);
CHECK(line == "# TYPE root_test_metric histogram");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="4",policy="default"} 1)");
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="1",policy="default"} 1)");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="8",policy="default"} 2)");
std::getline(output, line);
Expand Down Expand Up @@ -337,10 +336,8 @@ TEST_CASE("Histogram double metrics", "[metrics][histogram]")
h.update(8.000);
h.to_json(j["top"]);

CHECK(j["top"]["test"]["metric"]["buckets"]["+Inf"] == 4.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["12.000000"] == 4.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["4.000000"] == 0.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["8.000000"] == 1.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["12.915497"] == 4.0);
CHECK(j["top"]["test"]["metric"]["buckets"]["8.799225"] == 1.0);
}

SECTION("Histogram get n")
Expand All @@ -363,11 +360,11 @@ TEST_CASE("Histogram double metrics", "[metrics][histogram]")
std::getline(output, line);
CHECK(line == "# TYPE root_test_metric histogram");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="4.000033",policy="default"} 1)");
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="1.000000",policy="default"} 1)");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="8.000067",policy="default"} 2)");
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="8.799225",policy="default"} 2)");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="12.000100",policy="default"} 5)");
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="12.915497",policy="default"} 5)");
std::getline(output, line);
CHECK(line == R"(root_test_metric_bucket{instance="test instance",le="+Inf",policy="default"} 5)");
std::getline(output, line);
Expand Down