Enable speedb features: Constrain the i/f of SharedOptions (make immu…

…table)

HISTORY.md

@@ -18,6 +18,7 @@
 * Added a kUseBaseAddress flag and GetBaseOffset flag to OptionTypeInfo.  If this flag is set and a function is used for processing options, the function is passed the base address of the struct rather than the specific field (#397)
 * Static Pinning: Set the default for mid-percent capacity threshold in scoped pinning policy to 70 (#689).
 * db_bench: Add support for individual scoped pinning policy parameters (#687).
+* Enable speedb features: Constrain the interface of SharedOptions (make immutable) (#740).
 
 ### Bug Fixes
 * db_bench: Fix SeekRandomWriteRandom valid check. Use key and value only after checking iterator is valid.

examples/enable_speedb_features_example.cc

@@ -20,6 +20,7 @@
 #include "rocksdb/db.h"
 #include "rocksdb/options.h"
 #include "rocksdb/slice.h"
+#include "rocksdb/slice_transform.h"
 
 using namespace ROCKSDB_NAMESPACE;
 
@@ -55,11 +56,17 @@ int main() {
   // as listed in the definition of SpeedbSharedOptiopns in options.h
   op1.create_if_missing = true;
   op1.compression = rocksdb::kNoCompression;
+  // NOT Having a prefix-extractor (the deafult) will result in the
+  // memtable_factory==HashSpdbRepFactory
   //...
   op1.EnableSpeedbFeatures(so1);
 
   op2.create_if_missing = true;
   op2.compression = rocksdb::kZlibCompression;
+  // Having a prefix-extractor will result in the
+  // memtable_factory==SkipListRepFactory
+  op2.prefix_extractor.reset(NewFixedPrefixTransform(4));
+
   //...
   op2.EnableSpeedbFeatures(so1);
 
@@ -124,6 +131,8 @@ int main() {
   }
   std::cout << "new_cf was created in db3" << std::endl;
 
+  // Cleanup
+
   s = db3->DropColumnFamily(cf);
   if (!s.ok()) {
     std::cerr << s.ToString() << std::endl;

include/rocksdb/options.h

@@ -2241,39 +2241,72 @@ struct LiveFilesStorageInfoOptions {
 // more info and use example can be found in  enable_speedb_features_example.cc
 class SharedOptions {
  public:
-  SharedOptions();
+  static constexpr size_t kDefaultDelayedWriteRate = 256 * 1024 * 1024ul;
+  static constexpr size_t kDefaultBucketSize = 1000000;
+  static constexpr bool kDeafultUseMerge = true;
+
+  static constexpr size_t kWbmPerCfSizeIncrease = 512 * 1024 * 1024ul;
+
+ public:
   SharedOptions(size_t total_ram_size_bytes, size_t total_threads,
-                size_t delayed_write_rate = 256 * 1024 * 1024ul,
-                size_t bucket_size = 1000000, bool use_merge = true);
-  size_t GetTotalThreads() { return total_threads_; }
-  size_t GetTotalRamSizeBytes() { return total_ram_size_bytes_; }
-  size_t GetDelayedWriteRate() { return delayed_write_rate_; }
-  size_t GetBucketSize() { return bucket_size_; }
-  size_t IsMergeMemtableSupported() { return use_merge_; }
+                size_t delayed_write_rate = kDefaultDelayedWriteRate,
+                size_t bucket_size = kDefaultBucketSize,
+                bool use_merge = kDeafultUseMerge);
+
+ public:
+  size_t GetMaxWriteBufferManagerSize() const;
+
+  size_t GetTotalThreads() const { return total_threads_; }
+  size_t GetTotalRamSizeBytes() const { return total_ram_size_bytes_; }
+  size_t GetDelayedWriteRate() const { return delayed_write_rate_; }
+  size_t GetBucketSize() const { return bucket_size_; }
+  size_t IsMergeMemtableSupported() const { return use_merge_; }
+
+  const Cache* GetCache() const { return cache_.get(); }
+  const WriteController* GetWriteController() const {
+    return write_controller_.get();
+  };
+  const WriteBufferManager* GetWriteBufferManager() const {
+    return write_buffer_manager_.get();
+  }
+  const TablePinningPolicy* GetPinningPolicy() const {
+    return pinning_policy_.get();
+  }
+
+ private:
+  void CreateWriteBufferManager();
+  void CreatePinningPolicy();
+
   // this function will increase write buffer manager by increased_by amount
   // as long as the result is not bigger than the maximum size of
   // total_ram_size_ /4
   void IncreaseWriteBufferSize(size_t increase_by);
-  void CreatePinningPolicy();
-  size_t GetMaxWriteBufferManagerSize() const;
 
-  std::shared_ptr<Cache> cache = nullptr;
-  std::shared_ptr<WriteController> write_controller = nullptr;
-  std::shared_ptr<WriteBufferManager> write_buffer_manager = nullptr;
+ private:
+  std::shared_ptr<Cache> cache_ = nullptr;
+  std::shared_ptr<WriteController> write_controller_ = nullptr;
+  std::shared_ptr<WriteBufferManager> write_buffer_manager_ = nullptr;
+  std::shared_ptr<TablePinningPolicy> pinning_policy_ = nullptr;
+
+ private:
+  size_t total_ram_size_bytes_ = 0;
+  size_t total_threads_ = 0;
+  size_t delayed_write_rate_ = kDefaultBucketSize;
+  size_t bucket_size_ = kDefaultBucketSize;
+  bool use_merge_ = kDeafultUseMerge;
+
+ private:
+  // For Future Use
   Env* env = Env::Default();
   std::shared_ptr<RateLimiter> rate_limiter = nullptr;
   std::shared_ptr<SstFileManager> sst_file_manager = nullptr;
   std::shared_ptr<Logger> info_log = nullptr;
   std::vector<std::shared_ptr<EventListener>> listeners;
   std::shared_ptr<FileChecksumGenFactory> file_checksum_gen_factory = nullptr;
-  std::shared_ptr<TablePinningPolicy> pinning_policy = nullptr;
 
  private:
-  size_t total_threads_ = 0;
-  size_t total_ram_size_bytes_ = 0;
-  size_t delayed_write_rate_ = 0;
-  size_t bucket_size_ = 1000000;
-  bool use_merge_ = true;
+  friend struct DBOptions;
+  friend struct ColumnFamilyOptions;
 };
 
 }  // namespace ROCKSDB_NAMESPACE

options/options.cc

@@ -551,33 +551,32 @@ Options* Options::OldDefaults(int rocksdb_major_version,
 Options* Options::EnableSpeedbFeatures(SharedOptions& shared_options) {
   EnableSpeedbFeaturesDB(shared_options);
   EnableSpeedbFeaturesCF(shared_options);
+  if (memtable_factory->IsInsertConcurrentlySupported() == false) {
+    allow_concurrent_memtable_write = false;
+  }
   return this;
 }
 
 SharedOptions::SharedOptions(size_t total_ram_size_bytes, size_t total_threads,
                              size_t delayed_write_rate, size_t bucket_size,
-                             bool use_merge) {
-  total_threads_ = total_threads;
-  total_ram_size_bytes_ = total_ram_size_bytes;
-  delayed_write_rate_ = delayed_write_rate;
-  bucket_size_ = bucket_size;
-  use_merge_ = use_merge;
-  // initial_write_buffer_size_ is initialized to 1 to avoid from empty memory
-  // which might cause some problems
-  int initial_write_buffer_size_ = 1;
-  cache = NewLRUCache(total_ram_size_bytes_);
-  write_controller.reset(
+                             bool use_merge)
+    : total_ram_size_bytes_(total_ram_size_bytes),
+      total_threads_(total_threads),
+      delayed_write_rate_(delayed_write_rate),
+      bucket_size_(bucket_size),
+      use_merge_(use_merge) {
+  cache_ = NewLRUCache(total_ram_size_bytes_);
+  write_controller_.reset(
       new WriteController(true /*dynamic_delay*/, delayed_write_rate_));
-  write_buffer_manager.reset(new WriteBufferManager(
-      initial_write_buffer_size_, cache, true /*allow_stall*/));
 
+  CreateWriteBufferManager();
   CreatePinningPolicy();
 }
 
 void SharedOptions::IncreaseWriteBufferSize(size_t increase_by) {
   // Max write_buffer_manager->buffer_size()
   size_t wbm_max_buf_size = GetMaxWriteBufferManagerSize();
-  size_t current_buffer_size = write_buffer_manager->buffer_size();
+  size_t current_buffer_size = write_buffer_manager_->buffer_size();
   size_t set_buf_res = 0;
 
   if (current_buffer_size == 1 && increase_by > 1) {
@@ -591,14 +590,25 @@ void SharedOptions::IncreaseWriteBufferSize(size_t increase_by) {
     set_buf_res = wbm_max_buf_size;
   }
   if (set_buf_res != 0) {
-    write_buffer_manager->SetBufferSize(set_buf_res);
+    write_buffer_manager_->SetBufferSize(set_buf_res);
   }
 }
 
+void SharedOptions::CreateWriteBufferManager() {
+  // initial_write_buffer_size_ is initialized to 1 to avoid from empty memory
+  // which might cause some problems
+  size_t initial_write_buffer_size_ = 1U;
+
+  write_buffer_manager_.reset(new WriteBufferManager(
+      initial_write_buffer_size_, cache_, true /*allow_stall*/,
+      true /* initiate_fluses */, WriteBufferManager::FlushInitiationOptions(),
+      WriteBufferManager::kDfltStartDelayPercentThreshold));
+}
+
 void SharedOptions::CreatePinningPolicy() {
   // Calculate the size of the clean memory
-  auto clean_memory_capacity = cache->GetCapacity();
-  if (write_buffer_manager->cost_to_cache()) {
+  auto clean_memory_capacity = cache_->GetCapacity();
+  if (write_buffer_manager_->cost_to_cache()) {
     // The WBM's size is increased on every call to EnableSpeedbFeaturesCF()
     // up to a max size. For simplicity, calculate the space for pinning
     // as if wbm is at its max size. Otherwise we would have to update the
@@ -622,7 +632,7 @@ void SharedOptions::CreatePinningPolicy() {
   std::ostringstream oss;
   oss << "id=speedb_scoped_pinning_policy; capacity=" << pinning_capacity;
   auto s = TablePinningPolicy::CreateFromString(config_options, oss.str(),
-                                                &pinning_policy);
+                                                &pinning_policy_);
   assert(s.ok());
 }
 
@@ -631,13 +641,12 @@ size_t SharedOptions::GetMaxWriteBufferManagerSize() const {
 }
 
 DBOptions* DBOptions::EnableSpeedbFeaturesDB(SharedOptions& shared_options) {
-  env = shared_options.env;
   IncreaseParallelism((int)shared_options.GetTotalThreads());
   delayed_write_rate = shared_options.GetDelayedWriteRate();
   bytes_per_sync = 1ul << 20;
   use_dynamic_delay = true;
-  write_buffer_manager = shared_options.write_buffer_manager;
-  write_controller = shared_options.write_controller;
+  write_buffer_manager = shared_options.write_buffer_manager_;
+  write_controller = shared_options.write_controller_;
   return this;
 }
 
@@ -665,8 +674,8 @@ ColumnFamilyOptions* ColumnFamilyOptions::EnableSpeedbFeaturesCF(
   // to disable flush due to write buffer full
   // each new column family will ask the write buffer manager to increase the
   // write buffer size by 512 * 1024 * 1024ul
-  shared_options.IncreaseWriteBufferSize(512 * 1024 * 1024ul);
-  auto db_wbf_size = shared_options.write_buffer_manager->buffer_size();
+  shared_options.IncreaseWriteBufferSize(SharedOptions::kWbmPerCfSizeIncrease);
+  auto db_wbf_size = shared_options.write_buffer_manager_->buffer_size();
   // cf write_buffer_size
   write_buffer_size = std::min<size_t>(db_wbf_size / 4, 64ul << 20);
   max_write_buffer_number = 4;
@@ -682,15 +691,14 @@ ColumnFamilyOptions* ColumnFamilyOptions::EnableSpeedbFeaturesCF(
         &block_based_table_options.filter_policy);
     assert(s.ok());
     block_based_table_options.cache_index_and_filter_blocks = true;
-    block_based_table_options.block_cache = shared_options.cache;
+    block_based_table_options.block_cache = shared_options.cache_;
     block_based_table_options.cache_index_and_filter_blocks_with_high_priority =
         true;
-    block_based_table_options.pinning_policy = shared_options.pinning_policy;
+    block_based_table_options.pinning_policy = shared_options.pinning_policy_;
     table_factory.reset(NewBlockBasedTableFactory(block_based_table_options));
   }
   if (prefix_extractor) {
-    memtable_factory.reset(
-        NewHashSkipListRepFactory(shared_options.GetBucketSize()));
+    memtable_factory.reset(new SkipListFactory());
   } else {
     memtable_factory.reset(
         NewHashSpdbRepFactory(shared_options.GetBucketSize(),