Performant util/histogram.

Summary:
Earlier way to record in histogram=>
Linear search BucketLimit array to find the bucket and increment the
counter
Current way to record in histogram=>
Store a HistMap statically which points the buckets of each value in the
range [kFirstValue, kLastValue);

In the proccess use vectors instead of array's and refactor some code to
HistogramHelper class.

Test Plan:
run db_bench with histogram=1 and see a histogram being
printed.

Reviewers: dhruba, chip, heyongqiang

Reviewed By: chip

CC: leveldb

Differential Revision: https://reviews.facebook.net/D8265

Makefile

@@ -38,6 +38,7 @@ TESTS = \
 	c_test \
 	cache_test \
 	coding_test \
+	histogram_test \
 	corruption_test \
 	crc32c_test \
 	db_test \
@@ -149,6 +150,9 @@ cache_test: util/cache_test.o $(LIBOBJECTS) $(TESTHARNESS)
 coding_test: util/coding_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(CXX) util/coding_test.o $(LIBOBJECTS) $(TESTHARNESS) $(EXEC_LDFLAGS) -o$@ $(LDFLAGS)
 
+histogram_test: util/histogram_test.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(CXX) util/histogram_test.o $(LIBOBJECTS) $(TESTHARNESS) $(EXEC_LDFLAGS) -o$@ $(LDFLAGS)
+
 corruption_test: db/corruption_test.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(CXX) db/corruption_test.o $(LIBOBJECTS) $(TESTHARNESS) $(EXEC_LDFLAGS) -o$@ $(LDFLAGS)
 

util/histogram.cc

@@ -2,6 +2,7 @@
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file. See the AUTHORS file for names of contributors.
 
+#include <cassert>
 #include <math.h>
 #include <stdio.h>
 #include "port/port.h"
@@ -9,7 +10,10 @@
 
 namespace leveldb {
 
-const double Histogram::kBucketLimit[kNumBuckets] = {
+HistogramBucketMapper::HistogramBucketMapper() :
+  // Add newer bucket index here.
+  // Should be alwyas added in sorted order.
+  bucketValues_({
   1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16, 18, 20, 25, 30, 35, 40, 45,
   50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200, 250, 300, 350, 400, 450,
   500, 600, 700, 800, 900, 1000, 1200, 1400, 1600, 1800, 2000, 2500, 3000,
@@ -24,30 +28,50 @@ const double Histogram::kBucketLimit[kNumBuckets] = {
   70000000, 80000000, 90000000, 100000000, 120000000, 140000000, 160000000,
   180000000, 200000000, 250000000, 300000000, 350000000, 400000000,
   450000000, 500000000, 600000000, 700000000, 800000000, 900000000,
-  1000000000, 1200000000, 1400000000, 1600000000, 1800000000, 2000000000,
-  2500000000.0, 3000000000.0, 3500000000.0, 4000000000.0, 4500000000.0,
-  5000000000.0, 6000000000.0, 7000000000.0, 8000000000.0, 9000000000.0,
-  1e200,
-};
+  1000000000}),
+  maxBucketValue_(bucketValues_.back()),
+  minBucketValue_(bucketValues_.front()) {
+  for (size_t i =0; i < bucketValues_.size(); ++i) {
+    valueIndexMap_[bucketValues_[i]] = i;
+  }
+}
+
+const size_t HistogramBucketMapper::IndexForValue(const uint64_t value) const {
+  if (value >= maxBucketValue_) {
+    return bucketValues_.size() - 1;
+  } else if ( value >= minBucketValue_ ) {
+    std::map<uint64_t, uint64_t>::const_iterator lowerBound =
+      valueIndexMap_.lower_bound(value);
+    if (lowerBound != valueIndexMap_.end()) {
+      return lowerBound->second;
+    } else {
+      return 0;
+    }
+  } else {
+    return 0;
+  }
+}
+
+namespace {
+  const HistogramBucketMapper bucketMapper;
+}
+
+
+Histogram::Histogram() :
+  buckets_(std::vector<uint64_t>(bucketMapper.BucketCount(), 0)) {}
 
 void Histogram::Clear() {
-  min_ = kBucketLimit[kNumBuckets-1];
+  min_ = bucketMapper.LastValue();
   max_ = 0;
   num_ = 0;
   sum_ = 0;
   sum_squares_ = 0;
-  for (int i = 0; i < kNumBuckets; i++) {
-    buckets_[i] = 0;
-  }
+  buckets_.resize(bucketMapper.BucketCount(), 0);
 }
 
-void Histogram::Add(double value) {
-  // Linear search is fast enough for our usage in db_bench
-  int b = 0;
-  while (b < kNumBuckets - 1 && kBucketLimit[b] <= value) {
-    b++;
-  }
-  buckets_[b] += 1.0;
+void Histogram::Add(uint64_t value) {
+  const size_t index = bucketMapper.IndexForValue(value);
+  buckets_[index] += 1;
   if (min_ > value) min_ = value;
   if (max_ < value) max_ = value;
   num_++;
@@ -55,13 +79,17 @@ void Histogram::Add(double value) {
   sum_squares_ += (value * value);
 }
 
+void Histogram::Add(double value) {
+  Add(static_cast<uint64_t>(value));
+}
+
 void Histogram::Merge(const Histogram& other) {
   if (other.min_ < min_) min_ = other.min_;
   if (other.max_ > max_) max_ = other.max_;
   num_ += other.num_;
   sum_ += other.sum_;
   sum_squares_ += other.sum_squares_;
-  for (int b = 0; b < kNumBuckets; b++) {
+  for (int b = 0; b < bucketMapper.BucketCount(); b++) {
     buckets_[b] += other.buckets_[b];
   }
 }
@@ -73,15 +101,19 @@ double Histogram::Median() const {
 double Histogram::Percentile(double p) const {
   double threshold = num_ * (p / 100.0);
   double sum = 0;
-  for (int b = 0; b < kNumBuckets; b++) {
+  for (int b = 0; b < bucketMapper.BucketCount(); b++) {
     sum += buckets_[b];
     if (sum >= threshold) {
       // Scale linearly within this bucket
-      double left_point = (b == 0) ? 0 : kBucketLimit[b-1];
-      double right_point = kBucketLimit[b];
+      double left_point = (b == 0) ? 0 : bucketMapper.BucketLimit(b-1);
+      double right_point = bucketMapper.BucketLimit(b);
       double left_sum = sum - buckets_[b];
       double right_sum = sum;
-      double pos = (threshold - left_sum) / (right_sum - left_sum);
+      double pos = 0;
+      double right_left_diff = right_sum - left_sum;
+      if (right_left_diff != 0) {
+       pos = (threshold - left_sum) / (right_sum - left_sum);
+      }
       double r = left_point + (right_point - left_point) * pos;
       if (r < min_) r = min_;
       if (r > max_) r = max_;
@@ -116,13 +148,13 @@ std::string Histogram::ToString() const {
   r.append("------------------------------------------------------\n");
   const double mult = 100.0 / num_;
   double sum = 0;
-  for (int b = 0; b < kNumBuckets; b++) {
+  for (int b = 0; b < bucketMapper.BucketCount(); b++) {
     if (buckets_[b] <= 0.0) continue;
     sum += buckets_[b];
     snprintf(buf, sizeof(buf),
-             "[ %7.0f, %7.0f ) %7.0f %7.3f%% %7.3f%% ",
-             ((b == 0) ? 0.0 : kBucketLimit[b-1]),      // left
-             kBucketLimit[b],                           // right
+             "[ %ld, %ld ) %ld %7.3f%% %7.3f%% ",
+             ((b == 0) ? 0 : bucketMapper.BucketLimit(b-1)),      // left
+             bucketMapper.BucketLimit(b),                           // right
              buckets_[b],                             // count
              mult * buckets_[b],                      // percentage
              mult * sum);                             // cumulative percentage

util/histogram.h

@@ -5,36 +5,70 @@
 #ifndef STORAGE_LEVELDB_UTIL_HISTOGRAM_H_
 #define STORAGE_LEVELDB_UTIL_HISTOGRAM_H_
 
+#include <cassert>
 #include <string>
+#include <vector>
+#include <map>
 
 namespace leveldb {
 
+class HistogramBucketMapper {
+ public:
+
+  HistogramBucketMapper();
+
+  // converts a value to the bucket index.
+  const size_t IndexForValue(const uint64_t value) const;
+  // number of buckets required.
+
+  const size_t BucketCount() const {
+    return bucketValues_.size();
+  }
+
+  uint64_t LastValue() const {
+    return maxBucketValue_;
+  }
+
+  uint64_t FirstValue() const {
+    return minBucketValue_;
+  }
+
+  uint64_t BucketLimit(const uint64_t bucketNumber) const {
+    assert(bucketNumber < BucketCount());
+    return bucketValues_[bucketNumber];
+  }
+
+ private:
+  const std::vector<uint64_t> bucketValues_;
+  const uint64_t maxBucketValue_;
+  const uint64_t minBucketValue_;
+  std::map<uint64_t, uint64_t> valueIndexMap_;
+};
+
 class Histogram {
  public:
-  Histogram() { }
-  ~Histogram() { }
+  Histogram();
 
   void Clear();
+  void Add(uint64_t value);
   void Add(double value);
   void Merge(const Histogram& other);
 
   std::string ToString() const;
 
+  double Median() const;
+  double Percentile(double p) const;
+  double Average() const;
+  double StandardDeviation() const;
+
  private:
   double min_;
   double max_;
   double num_;
   double sum_;
   double sum_squares_;
+  std::vector<uint64_t> buckets_;
 
-  enum { kNumBuckets = 154 };
-  static const double kBucketLimit[kNumBuckets];
-  double buckets_[kNumBuckets];
-
-  double Median() const;
-  double Percentile(double p) const;
-  double Average() const;
-  double StandardDeviation() const;
 };
 
 }  // namespace leveldb

util/histogram_test.cc

@@ -0,0 +1,57 @@
+#include "util/histogram.h"
+
+#include "util/testharness.h"
+
+namespace leveldb {
+
+class HistogramTest { };
+
+TEST(HistogramTest, BasicOperation) {
+
+  Histogram histogram;
+  for (uint64_t i = 1; i <= 100; i++) {
+    histogram.Add(i);
+  }
+
+  {
+    double median = histogram.Median();
+    // ASSERT_LE(median, 50);
+    ASSERT_GT(median, 0);
+  }
+
+  {
+    double percentile100 = histogram.Percentile(100.0);
+    ASSERT_LE(percentile100, 100.0);
+    ASSERT_GT(percentile100, 0.0);
+    double percentile99 = histogram.Percentile(99.0);
+    double percentile85 = histogram.Percentile(85.0);
+    ASSERT_LE(percentile99, 99.0);
+    ASSERT_TRUE(percentile99 >= percentile85);
+  }
+
+  ASSERT_EQ(histogram.Average(), 50.5); // avg is acurately caluclated.
+}
+
+TEST(HistogramTest, EmptyHistogram) {
+  Histogram histogram;
+  ASSERT_EQ(histogram.Median(), 0.0);
+  ASSERT_EQ(histogram.Percentile(85.0), 0.0);
+  ASSERT_EQ(histogram.Average(), 0.0);
+}
+
+TEST(HistogramTest, ClearHistogram) {
+  Histogram histogram;
+  for (uint64_t i = 1; i <= 100; i++) {
+    histogram.Add(i);
+  }
+  histogram.Clear();
+  ASSERT_EQ(histogram.Median(), 0);
+  ASSERT_EQ(histogram.Percentile(85.0), 0);
+  ASSERT_EQ(histogram.Average(), 0);
+}
+
+}  // namespace leveldb
+
+int main(int argc, char** argv) {
+  return leveldb::test::RunAllTests();
+}