Implementation for GetPropertiesOfTablesInRange

Summary: In MyRocks, it is sometimes important to get propeties only for the subset of the database. This diff implements the API in RocksDB.

Test Plan: ran the GetPropertiesOfTablesInRange

Reviewers: rven, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D48651

CMakeLists.txt

@@ -294,6 +294,7 @@ set(TESTS
         db/db_universal_compaction_test.cc
         db/db_wal_test.cc
         db/db_tailing_iter_test.cc
+        db/db_table_properties_test.cc
         db/dbformat_test.cc
         db/deletefile_test.cc
         db/fault_injection_test.cc

Makefile

@@ -230,6 +230,7 @@ TESTS = \
 	db_tailing_iter_test \
 	db_universal_compaction_test \
 	db_wal_test \
+	db_table_properties_test \
 	block_hash_index_test \
 	autovector_test \
 	column_family_test \
@@ -741,6 +742,9 @@ db_universal_compaction_test: db/db_universal_compaction_test.o db/db_test_util.
 db_wal_test: db/db_wal_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK)
 
+db_table_properties_test: db/db_table_properties_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
+	$(AM_LINK)
+
 log_write_bench: util/log_write_bench.o $(LIBOBJECTS) $(TESTHARNESS)
 	$(AM_LINK) $(pg)
 

db/db_impl.cc

@@ -4161,6 +4161,29 @@ Status DBImpl::GetPropertiesOfAllTables(ColumnFamilyHandle* column_family,
 
   return s;
 }
+
+Status DBImpl::GetPropertiesOfTablesInRange(ColumnFamilyHandle* column_family,
+                                            const Range* range, int n,
+                                            TablePropertiesCollection* props) {
+  auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
+  auto cfd = cfh->cfd();
+
+  // Increment the ref count
+  mutex_.Lock();
+  auto version = cfd->current();
+  version->Ref();
+  mutex_.Unlock();
+
+  auto s = version->GetPropertiesOfTablesInRange(range, n, props);
+
+  // Decrement the ref count
+  mutex_.Lock();
+  version->Unref();
+  mutex_.Unlock();
+
+  return s;
+}
+
 #endif  // ROCKSDB_LITE
 
 const std::string& DBImpl::GetName() const {

db/db_impl.h

@@ -810,6 +810,10 @@ class DBImpl : public DB {
   virtual Status GetPropertiesOfAllTables(ColumnFamilyHandle* column_family,
                                           TablePropertiesCollection* props)
       override;
+  virtual Status GetPropertiesOfTablesInRange(
+      ColumnFamilyHandle* column_family, const Range* range, int n,
+      TablePropertiesCollection* props) override;
+
 #endif  // ROCKSDB_LITE
 
   // Function that Get and KeyMayExist call with no_io true or false

db/db_table_properties_test.cc

@@ -0,0 +1,217 @@
+//  Copyright (c) 2013, Facebook, Inc.  All rights reserved.
+//  This source code is licensed under the BSD-style license found in the
+//  LICENSE file in the root directory of this source tree. An additional grant
+//  of patent rights can be found in the PATENTS file in the same directory.
+//
+// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
+// 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 <unordered_set>
+#include <vector>
+
+#include "db/db_test_util.h"
+#include "port/stack_trace.h"
+#include "rocksdb/db.h"
+#include "util/testharness.h"
+#include "util/testutil.h"
+
+#ifndef ROCKSDB_LITE
+
+namespace rocksdb {
+
+// A helper function that ensures the table properties returned in
+// `GetPropertiesOfAllTablesTest` is correct.
+// This test assumes entries size is different for each of the tables.
+namespace {
+
+void VerifyTableProperties(DB* db, uint64_t expected_entries_size) {
+  TablePropertiesCollection props;
+  ASSERT_OK(db->GetPropertiesOfAllTables(&props));
+
+  ASSERT_EQ(4U, props.size());
+  std::unordered_set<uint64_t> unique_entries;
+
+  // Indirect test
+  uint64_t sum = 0;
+  for (const auto& item : props) {
+    unique_entries.insert(item.second->num_entries);
+    sum += item.second->num_entries;
+  }
+
+  ASSERT_EQ(props.size(), unique_entries.size());
+  ASSERT_EQ(expected_entries_size, sum);
+}
+}  // namespace
+
+class DBTablePropertiesTest : public DBTestBase {
+ public:
+  DBTablePropertiesTest() : DBTestBase("/db_table_properties_test") {}
+  TablePropertiesCollection TestGetPropertiesOfTablesInRange(
+      std::vector<Range> ranges, std::size_t* num_properties = nullptr,
+      std::size_t* num_files = nullptr);
+};
+
+TEST_F(DBTablePropertiesTest, GetPropertiesOfAllTablesTest) {
+  Options options = CurrentOptions();
+  options.level0_file_num_compaction_trigger = 8;
+  Reopen(options);
+  // Create 4 tables
+  for (int table = 0; table < 4; ++table) {
+    for (int i = 0; i < 10 + table; ++i) {
+      db_->Put(WriteOptions(), ToString(table * 100 + i), "val");
+    }
+    db_->Flush(FlushOptions());
+  }
+
+  // 1. Read table properties directly from file
+  Reopen(options);
+  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
+
+  // 2. Put two tables to table cache and
+  Reopen(options);
+  // fetch key from 1st and 2nd table, which will internally place that table to
+  // the table cache.
+  for (int i = 0; i < 2; ++i) {
+    Get(ToString(i * 100 + 0));
+  }
+
+  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
+
+  // 3. Put all tables to table cache
+  Reopen(options);
+  // fetch key from 1st and 2nd table, which will internally place that table to
+  // the table cache.
+  for (int i = 0; i < 4; ++i) {
+    Get(ToString(i * 100 + 0));
+  }
+  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
+}
+
+TablePropertiesCollection
+DBTablePropertiesTest::TestGetPropertiesOfTablesInRange(
+    std::vector<Range> ranges, std::size_t* num_properties,
+    std::size_t* num_files) {
+  // run the query
+  TablePropertiesCollection props;
+  EXPECT_OK(db_->GetPropertiesOfTablesInRange(
+      db_->DefaultColumnFamily(), &ranges[0], ranges.size(), &props));
+
+  // Make sure that we've received properties for those and for those files
+  // only which fall within requested ranges
+  std::vector<LiveFileMetaData> vmd;
+  db_->GetLiveFilesMetaData(&vmd);
+  for (auto md : vmd) {
+    std::string fn = md.db_path + md.name;
+    bool in_range = false;
+    for (auto r : ranges) {
+      // smallestkey < limit && largestkey >= start
+      if (r.limit.compare(md.smallestkey) >= 0 &&
+          r.start.compare(md.largestkey) <= 0) {
+        in_range = true;
+        EXPECT_GT(props.count(fn), 0);
+      }
+    }
+    if (!in_range) {
+      EXPECT_EQ(props.count(fn), 0);
+    }
+  }
+
+  if (num_properties) {
+    *num_properties = props.size();
+  }
+
+  if (num_files) {
+    *num_files = vmd.size();
+  }
+  return props;
+}
+
+TEST_F(DBTablePropertiesTest, GetPropertiesOfTablesInRange) {
+  // Fixed random sead
+  Random rnd(301);
+
+  Options options;
+  options.create_if_missing = true;
+  options.write_buffer_size = 4096;
+  options.max_write_buffer_number = 8;
+  options.level0_file_num_compaction_trigger = 2;
+  options.level0_slowdown_writes_trigger = 2;
+  options.level0_stop_writes_trigger = 4;
+  options.target_file_size_base = 2048;
+  options.max_bytes_for_level_base = 10240;
+  options.max_bytes_for_level_multiplier = 4;
+  options.soft_rate_limit = 1.1;
+  options.num_levels = 8;
+
+  DestroyAndReopen(options);
+
+  // build a decent LSM
+  for (int i = 0; i < 10000; i++) {
+    ASSERT_OK(Put(test::RandomKey(&rnd, 5), RandomString(&rnd, 102)));
+  }
+  Flush();
+  db_->PauseBackgroundWork();
+
+  // Ensure that we have at least L0, L1 and L2
+  ASSERT_GT(NumTableFilesAtLevel(0), 0);
+  ASSERT_GT(NumTableFilesAtLevel(1), 0);
+  ASSERT_GT(NumTableFilesAtLevel(2), 0);
+
+  // Query the largest range
+  std::size_t num_properties, num_files;
+  TestGetPropertiesOfTablesInRange(
+      {Range(test::RandomKey(&rnd, 5, test::RandomKeyType::SMALLEST),
+             test::RandomKey(&rnd, 5, test::RandomKeyType::LARGEST))},
+      &num_properties, &num_files);
+  ASSERT_EQ(num_properties, num_files);
+
+  // Query the empty range
+  TestGetPropertiesOfTablesInRange(
+      {Range(test::RandomKey(&rnd, 5, test::RandomKeyType::LARGEST),
+             test::RandomKey(&rnd, 5, test::RandomKeyType::SMALLEST))},
+      &num_properties, &num_files);
+  ASSERT_GT(num_files, 0);
+  ASSERT_EQ(num_properties, 0);
+
+  // Query the middle rangee
+  TestGetPropertiesOfTablesInRange(
+      {Range(test::RandomKey(&rnd, 5, test::RandomKeyType::MIDDLE),
+             test::RandomKey(&rnd, 5, test::RandomKeyType::LARGEST))},
+      &num_properties, &num_files);
+  ASSERT_GT(num_files, 0);
+  ASSERT_GT(num_files, num_properties);
+  ASSERT_GT(num_properties, 0);
+
+  // Query a bunch of random ranges
+  for (int j = 0; j < 100; j++) {
+    // create a bunch of ranges
+    std::vector<std::string> random_keys;
+    auto n = 2 * rnd.Uniform(50);
+    for (uint i = 0; i < n; ++i) {
+      random_keys.push_back(test::RandomKey(&rnd, 5));
+    }
+
+    std::vector<Range> ranges;
+    auto it = random_keys.begin();
+    while (it != random_keys.end()) {
+      ranges.push_back(Range(*it, *(it + 1)));
+      it += 2;
+    }
+
+    TestGetPropertiesOfTablesInRange(std::move(ranges));
+  }
+}
+}  // namespace rocksdb
+
+#endif  // ROCKSDB_LITE
+
+int main(int argc, char** argv) {
+#if !(defined NDEBUG) || !defined(OS_WIN)
+  rocksdb::port::InstallStackTraceHandler();
+  ::testing::InitGoogleTest(&argc, argv);
+  return RUN_ALL_TESTS();
+#else
+  return 0;
+#endif
+}

db/db_test.cc

@@ -84,24 +84,6 @@ static long TestGetTickerCount(const Options& options, Tickers ticker_type) {
 // This test assumes entries size is different for each of the tables.
 namespace {
 
-void VerifyTableProperties(DB* db, uint64_t expected_entries_size) {
-  TablePropertiesCollection props;
-  ASSERT_OK(db->GetPropertiesOfAllTables(&props));
-
-  ASSERT_EQ(4U, props.size());
-  std::unordered_set<uint64_t> unique_entries;
-
-  // Indirect test
-  uint64_t sum = 0;
-  for (const auto& item : props) {
-    unique_entries.insert(item.second->num_entries);
-    sum += item.second->num_entries;
-  }
-
-  ASSERT_EQ(props.size(), unique_entries.size());
-  ASSERT_EQ(expected_entries_size, sum);
-}
-
 uint64_t GetNumberOfSstFilesForColumnFamily(DB* db,
                                             std::string column_family_name) {
   std::vector<LiveFileMetaData> metadata;
@@ -440,42 +422,6 @@ TEST_F(DBTest, ParanoidFileChecks) {
             TestGetTickerCount(options, BLOCK_CACHE_ADD));
 }
 
-TEST_F(DBTest, GetPropertiesOfAllTablesTest) {
-  Options options = CurrentOptions();
-  options.level0_file_num_compaction_trigger = 8;
-  Reopen(options);
-  // Create 4 tables
-  for (int table = 0; table < 4; ++table) {
-    for (int i = 0; i < 10 + table; ++i) {
-      db_->Put(WriteOptions(), ToString(table * 100 + i), "val");
-    }
-    db_->Flush(FlushOptions());
-  }
-
-  // 1. Read table properties directly from file
-  Reopen(options);
-  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
-
-  // 2. Put two tables to table cache and
-  Reopen(options);
-  // fetch key from 1st and 2nd table, which will internally place that table to
-  // the table cache.
-  for (int i = 0; i < 2; ++i) {
-    Get(ToString(i * 100 + 0));
-  }
-
-  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
-
-  // 3. Put all tables to table cache
-  Reopen(options);
-  // fetch key from 1st and 2nd table, which will internally place that table to
-  // the table cache.
-  for (int i = 0; i < 4; ++i) {
-    Get(ToString(i * 100 + 0));
-  }
-  VerifyTableProperties(db_, 10 + 11 + 12 + 13);
-}
-
 namespace {
 void ResetTableProperties(TableProperties* tp) {
   tp->data_size = 0;
@@ -5692,6 +5638,12 @@ class ModelDB: public DB {
       TablePropertiesCollection* props) override {
     return Status();
   }
+
+  virtual Status GetPropertiesOfTablesInRange(
+      ColumnFamilyHandle* column_family, const Range* range, int n,
+      TablePropertiesCollection* props) override {
+    return Status();
+  }
 #endif  // ROCKSDB_LITE
 
   using DB::KeyMayExist;

db/version_set.cc

@@ -542,7 +542,7 @@ class BaseReferencedVersionBuilder {
 
 Status Version::GetTableProperties(std::shared_ptr<const TableProperties>* tp,
                                    const FileMetaData* file_meta,
-                                   const std::string* fname) {
+                                   const std::string* fname) const {
   auto table_cache = cfd_->table_cache();
   auto ioptions = cfd_->ioptions();
   Status s = table_cache->GetTableProperties(
@@ -624,6 +624,38 @@ Status Version::GetPropertiesOfAllTables(TablePropertiesCollection* props,
   return Status::OK();
 }
 
+Status Version::GetPropertiesOfTablesInRange(
+    const Range* range, int n, TablePropertiesCollection* props) const {
+  for (int level = 0; level < storage_info_.num_non_empty_levels(); level++) {
+    for (int i = 0; i < n; i++) {
+      // Convert user_key into a corresponding internal key.
+      InternalKey k1(range[i].start, kMaxSequenceNumber, kValueTypeForSeek);
+      InternalKey k2(range[i].limit, kMaxSequenceNumber, kValueTypeForSeek);
+      std::vector<FileMetaData*> files;
+      storage_info_.GetOverlappingInputs(level, &k1, &k2, &files, -1, nullptr,
+                                         false);
+      for (const auto& file_meta : files) {
+        auto fname =
+            TableFileName(vset_->db_options_->db_paths,
+                          file_meta->fd.GetNumber(), file_meta->fd.GetPathId());
+        if (props->count(fname) == 0) {
+          // 1. If the table is already present in table cache, load table
+          // properties from there.
+          std::shared_ptr<const TableProperties> table_properties;
+          Status s = GetTableProperties(&table_properties, file_meta, &fname);
+          if (s.ok()) {
+            props->insert({fname, table_properties});
+          } else {
+            return s;
+          }
+        }
+      }
+    }
+  }
+
+  return Status::OK();
+}
+
 Status Version::GetAggregatedTableProperties(
     std::shared_ptr<const TableProperties>* tp, int level) {
   TablePropertiesCollection props;
@@ -1406,7 +1438,8 @@ bool VersionStorageInfo::OverlapInLevel(int level,
 // The file_index returns a pointer to any file in an overlapping range.
 void VersionStorageInfo::GetOverlappingInputs(
     int level, const InternalKey* begin, const InternalKey* end,
-    std::vector<FileMetaData*>* inputs, int hint_index, int* file_index) {
+    std::vector<FileMetaData*>* inputs, int hint_index, int* file_index,
+    bool expand_range) const {
   if (level >= num_non_empty_levels_) {
     // this level is empty, no overlapping inputs
     return;
@@ -1439,7 +1472,7 @@ void VersionStorageInfo::GetOverlappingInputs(
       // "f" is completely after specified range; skip it
     } else {
       inputs->push_back(files_[level][i-1]);
-      if (level == 0) {
+      if (level == 0 && expand_range) {
         // Level-0 files may overlap each other.  So check if the newly
         // added file has expanded the range.  If so, restart search.
         if (begin != nullptr && user_cmp->Compare(file_start, user_begin) < 0) {
@@ -1465,7 +1498,7 @@ void VersionStorageInfo::GetOverlappingInputs(
 // forwards to find all overlapping files.
 void VersionStorageInfo::GetOverlappingInputsBinarySearch(
     int level, const Slice& user_begin, const Slice& user_end,
-    std::vector<FileMetaData*>* inputs, int hint_index, int* file_index) {
+    std::vector<FileMetaData*>* inputs, int hint_index, int* file_index) const {
   assert(level > 0);
   int min = 0;
   int mid = 0;
@@ -1513,8 +1546,7 @@ void VersionStorageInfo::GetOverlappingInputsBinarySearch(
 // Use FileLevel in searching, make it faster
 void VersionStorageInfo::ExtendOverlappingInputs(
     int level, const Slice& user_begin, const Slice& user_end,
-    std::vector<FileMetaData*>* inputs, unsigned int midIndex) {
-
+    std::vector<FileMetaData*>* inputs, unsigned int midIndex) const {
   const Comparator* user_cmp = user_comparator_;
   const FdWithKeyRange* files = level_files_brief_[level].files;
 #ifndef NDEBUG

db/version_set.h

@@ -159,23 +159,26 @@ class VersionStorageInfo {
       int level, const InternalKey* begin,  // nullptr means before all keys
       const InternalKey* end,               // nullptr means after all keys
       std::vector<FileMetaData*>* inputs,
-      int hint_index = -1,         // index of overlap file
-      int* file_index = nullptr);  // return index of overlap file
+      int hint_index = -1,        // index of overlap file
+      int* file_index = nullptr,  // return index of overlap file
+      bool expand_range = true)   // if set, returns files which overlap the
+      const;                      // range and overlap each other. If false,
+                                  // then just files intersecting the range
 
   void GetOverlappingInputsBinarySearch(
       int level,
       const Slice& begin,  // nullptr means before all keys
       const Slice& end,    // nullptr means after all keys
       std::vector<FileMetaData*>* inputs,
-      int hint_index,    // index of overlap file
-      int* file_index);  // return index of overlap file
+      int hint_index,          // index of overlap file
+      int* file_index) const;  // return index of overlap file
 
   void ExtendOverlappingInputs(
       int level,
       const Slice& begin,  // nullptr means before all keys
       const Slice& end,    // nullptr means after all keys
       std::vector<FileMetaData*>* inputs,
-      unsigned int index);  // start extending from this index
+      unsigned int index) const;  // start extending from this index
 
   // Returns true iff some file in the specified level overlaps
   // some part of [*smallest_user_key,*largest_user_key].
@@ -456,15 +459,16 @@ class Version {
   // file-name conversion.
   Status GetTableProperties(std::shared_ptr<const TableProperties>* tp,
                             const FileMetaData* file_meta,
-                            const std::string* fname = nullptr);
+                            const std::string* fname = nullptr) const;
 
   // REQUIRES: lock is held
   // On success, *props will be populated with all SSTables' table properties.
   // The keys of `props` are the sst file name, the values of `props` are the
   // tables' propertis, represented as shared_ptr.
   Status GetPropertiesOfAllTables(TablePropertiesCollection* props);
-
   Status GetPropertiesOfAllTables(TablePropertiesCollection* props, int level);
+  Status GetPropertiesOfTablesInRange(const Range* range, int n,
+                                      TablePropertiesCollection* props) const;
 
   // REQUIRES: lock is held
   // On success, "tp" will contains the aggregated table property amoug

include/rocksdb/db.h

@@ -716,6 +716,9 @@ class DB {
   virtual Status GetPropertiesOfAllTables(TablePropertiesCollection* props) {
     return GetPropertiesOfAllTables(DefaultColumnFamily(), props);
   }
+  virtual Status GetPropertiesOfTablesInRange(
+      ColumnFamilyHandle* column_family, const Range* range, int n,
+      TablePropertiesCollection* props) = 0;
 #endif  // ROCKSDB_LITE
 
   // Needed for StackableDB

include/rocksdb/utilities/stackable_db.h

@@ -279,6 +279,13 @@ class StackableDB : public DB {
     return db_->GetPropertiesOfAllTables(column_family, props);
   }
 
+  using DB::GetPropertiesOfTablesInRange;
+  virtual Status GetPropertiesOfTablesInRange(
+      ColumnFamilyHandle* column_family, const Range* range, int n,
+      TablePropertiesCollection* props) override {
+    return db_->GetPropertiesOfTablesInRange(column_family, range, n, props);
+  }
+
   virtual Status GetUpdatesSince(
       SequenceNumber seq_number, unique_ptr<TransactionLogIterator>* iter,
       const TransactionLogIterator::ReadOptions& read_options) override {

src.mk

@@ -190,6 +190,7 @@ TEST_BENCH_SOURCES =                                                    \
   db/db_universal_compaction_test.cc                                    \
   db/db_tailing_iter_test.cc                                            \
   db/db_wal_test.cc                                                     \
+  db/db_table_properties_test.cc                                        \
   db/deletefile_test.cc                                                 \
   db/fault_injection_test.cc                                            \
   db/file_indexer_test.cc                                               \

util/testutil.cc

@@ -33,7 +33,7 @@ extern std::string RandomHumanReadableString(Random* rnd, int len) {
   return ret;
 }
 
-std::string RandomKey(Random* rnd, int len) {
+std::string RandomKey(Random* rnd, int len, RandomKeyType type) {
   // Make sure to generate a wide variety of characters so we
   // test the boundary conditions for short-key optimizations.
   static const char kTestChars[] = {
@@ -41,7 +41,22 @@ std::string RandomKey(Random* rnd, int len) {
   };
   std::string result;
   for (int i = 0; i < len; i++) {
-    result += kTestChars[rnd->Uniform(sizeof(kTestChars))];
+    std::size_t indx = 0;
+    switch (type) {
+      case RandomKeyType::RANDOM:
+        indx = rnd->Uniform(sizeof(kTestChars));
+        break;
+      case RandomKeyType::LARGEST:
+        indx = sizeof(kTestChars) - 1;
+        break;
+      case RandomKeyType::MIDDLE:
+        indx = sizeof(kTestChars) / 2;
+        break;
+      case RandomKeyType::SMALLEST:
+        indx = 0;
+        break;
+    }
+    result += kTestChars[indx];
   }
   return result;
 }

util/testutil.h

@@ -35,7 +35,9 @@ extern std::string RandomHumanReadableString(Random* rnd, int len);
 
 // Return a random key with the specified length that may contain interesting
 // characters (e.g. \x00, \xff, etc.).
-extern std::string RandomKey(Random* rnd, int len);
+enum RandomKeyType : char { RANDOM, LARGEST, SMALLEST, MIDDLE };
+extern std::string RandomKey(Random* rnd, int len,
+                             RandomKeyType type = RandomKeyType::RANDOM);
 
 // Store in *dst a string of length "len" that will compress to
 // "N*compressed_fraction" bytes and return a Slice that references