Not scheduling more L1->L2 compaction if L0->L1 is pending with higher priority

Summary: When L0->L1 is pending, there may be one L1->L2 compaction going on which prevents the L0->L1 compaction from happening. If L1 needs more data to be moved to L2, then we may continue scheduling more L1->L2 compactions. The end result may be that L0->L1 compaction will not happen until L1 size drops to below target size. We can reduce the stalling because of number of L0 files by stopping schedling new L1->L2 compaction when L0's score is higher than L1.

Test Plan: Run all existing tests.

Reviewers: yhchiang, MarkCallaghan, rven, anthony, IslamAbdelRahman, igor

Reviewed By: igor

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52401

db/compaction_picker.cc

@@ -919,11 +919,18 @@ Compaction* LevelCompactionPicker::PickCompaction(
   CompactionReason compaction_reason = CompactionReason::kUnknown;
 
   // Find the compactions by size on all levels.
+  bool skipped_l0 = false;
   for (int i = 0; i < NumberLevels() - 1; i++) {
     score = vstorage->CompactionScore(i);
     level = vstorage->CompactionScoreLevel(i);
     assert(i == 0 || score <= vstorage->CompactionScore(i - 1));
     if (score >= 1) {
+      if (skipped_l0 && level == vstorage->base_level()) {
+        // If L0->base_level compaction is pending, don't schedule further
+        // compaction from base level. Otherwise L0->base_level compaction
+        // may starve.
+        continue;
+      }
       output_level = (level == 0) ? vstorage->base_level() : level + 1;
       if (PickCompactionBySize(vstorage, level, output_level, &inputs,
                                &parent_index, &base_index) &&
@@ -940,6 +947,9 @@ Compaction* LevelCompactionPicker::PickCompaction(
       } else {
         // didn't find the compaction, clear the inputs
         inputs.clear();
+        if (level == 0) {
+          skipped_l0 = true;
+        }
       }
     }
   }

db/compaction_picker_test.cc

@@ -580,6 +580,101 @@ TEST_F(CompactionPickerTest, OverlappingUserKeys3) {
   ASSERT_EQ(7U, compaction->input(1, 1)->fd.GetNumber());
 }
 
+TEST_F(CompactionPickerTest, NotScheduleL1IfL0WithHigherPri1) {
+  NewVersionStorage(6, kCompactionStyleLevel);
+  mutable_cf_options_.level0_file_num_compaction_trigger = 2;
+  mutable_cf_options_.max_bytes_for_level_base = 900000000U;
+
+  // 6 L0 files, score 3.
+  Add(0, 1U, "000", "400", 1U);
+  Add(0, 2U, "001", "400", 1U, 0, 0);
+  Add(0, 3U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 31U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 32U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 33U, "001", "400", 1000000000U, 0, 0);
+
+  // L1 total size 2GB, score 2.2. If one file being comapcted, score 1.1.
+  Add(1, 4U, "050", "300", 1000000000U, 0, 0);
+  file_map_[4u].first->being_compacted = true;
+  Add(1, 5U, "301", "350", 1000000000U, 0, 0);
+
+  // Output level overlaps with the beginning and the end of the chain
+  Add(2, 6U, "050", "100", 1U);
+  Add(2, 7U, "300", "400", 1U);
+
+  // No compaction should be scheduled, if L0 has higher priority than L1
+  // but L0->L1 compaction is blocked by a file in L1 being compacted.
+  UpdateVersionStorageInfo();
+  ASSERT_EQ(0, vstorage_->CompactionScoreLevel(0));
+  ASSERT_EQ(1, vstorage_->CompactionScoreLevel(1));
+  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
+      cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
+  ASSERT_TRUE(compaction.get() == nullptr);
+}
+
+TEST_F(CompactionPickerTest, NotScheduleL1IfL0WithHigherPri2) {
+  NewVersionStorage(6, kCompactionStyleLevel);
+  mutable_cf_options_.level0_file_num_compaction_trigger = 2;
+  mutable_cf_options_.max_bytes_for_level_base = 900000000U;
+
+  // 6 L0 files, score 3.
+  Add(0, 1U, "000", "400", 1U);
+  Add(0, 2U, "001", "400", 1U, 0, 0);
+  Add(0, 3U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 31U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 32U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 33U, "001", "400", 1000000000U, 0, 0);
+
+  // L1 total size 2GB, score 2.2. If one file being comapcted, score 1.1.
+  Add(1, 4U, "050", "300", 1000000000U, 0, 0);
+  Add(1, 5U, "301", "350", 1000000000U, 0, 0);
+
+  // Output level overlaps with the beginning and the end of the chain
+  Add(2, 6U, "050", "100", 1U);
+  Add(2, 7U, "300", "400", 1U);
+
+  // If no file in L1 being compacted, L0->L1 compaction will be scheduled.
+  UpdateVersionStorageInfo();  // being_compacted flag is cleared here.
+  ASSERT_EQ(0, vstorage_->CompactionScoreLevel(0));
+  ASSERT_EQ(1, vstorage_->CompactionScoreLevel(1));
+  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
+      cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
+  ASSERT_TRUE(compaction.get() != nullptr);
+}
+
+TEST_F(CompactionPickerTest, NotScheduleL1IfL0WithHigherPri3) {
+  NewVersionStorage(6, kCompactionStyleLevel);
+  mutable_cf_options_.level0_file_num_compaction_trigger = 2;
+  mutable_cf_options_.max_bytes_for_level_base = 900000000U;
+
+  // 6 L0 files, score 3.
+  Add(0, 1U, "000", "400", 1U);
+  Add(0, 2U, "001", "400", 1U, 0, 0);
+  Add(0, 3U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 31U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 32U, "001", "400", 1000000000U, 0, 0);
+  Add(0, 33U, "001", "400", 1000000000U, 0, 0);
+
+  // L1 score more than 6.
+  Add(1, 4U, "050", "300", 1000000000U, 0, 0);
+  file_map_[4u].first->being_compacted = true;
+  Add(1, 5U, "301", "350", 1000000000U, 0, 0);
+  Add(1, 51U, "351", "400", 6000000000U, 0, 0);
+
+  // Output level overlaps with the beginning and the end of the chain
+  Add(2, 6U, "050", "100", 1U);
+  Add(2, 7U, "300", "400", 1U);
+
+  // If score in L1 is larger than L0, L1 compaction goes through despite
+  // there is pending L0 compaction.
+  UpdateVersionStorageInfo();
+  ASSERT_EQ(1, vstorage_->CompactionScoreLevel(0));
+  ASSERT_EQ(0, vstorage_->CompactionScoreLevel(1));
+  std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
+      cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
+  ASSERT_TRUE(compaction.get() != nullptr);
+}
+
 TEST_F(CompactionPickerTest, EstimateCompactionBytesNeeded1) {
   int num_levels = ioptions_.num_levels;
   ioptions_.level_compaction_dynamic_level_bytes = false;