Ability to set different size fanout multipliers for every level.

Summary:
There is an existing field Options.max_bytes_for_level_multiplier that
sets the multiplier for the size of each level in the database.

This patch introduces the ability to set different multipliers
for every level in the database. The size of a level is determined
by using both max_bytes_for_level_multiplier as well as the
per-level fanout.

size of level[i] = size of level[i-1] * max_bytes_for_level_multiplier
                   * fanout[i-1]

The default value of fanout is 1, so that it is backward compatible.

Test Plan: make check

Reviewers: haobo, emayanke

Reviewed By: emayanke

CC: leveldb

Differential Revision: https://reviews.facebook.net/D10863

db/db_bench.cc

@@ -20,6 +20,7 @@
 #include "util/mutexlock.h"
 #include "util/random.h"
 #include "util/stack_trace.h"
+#include "util/string_util.h"
 #include "util/testutil.h"
 #include "hdfs/env_hdfs.h"
 
@@ -202,6 +203,9 @@ static uint64_t FLAGS_max_bytes_for_level_base = 10 * 1048576;
 // A multiplier to compute max bytes for level-N
 static int FLAGS_max_bytes_for_level_multiplier = 10;
 
+// A vector that specifies additional fanout per level
+static std::vector<int> FLAGS_max_bytes_for_level_multiplier_additional;
+
 // Number of files in level-0 that will trigger put stop.
 static int FLAGS_level0_stop_writes_trigger = 12;
 
@@ -1089,6 +1093,16 @@ unique_ptr<char []> GenerateKeyFromInt(int v, const char* suffix = "")
     options.max_bytes_for_level_base = FLAGS_max_bytes_for_level_base;
     options.max_bytes_for_level_multiplier =
         FLAGS_max_bytes_for_level_multiplier;
+    if (FLAGS_max_bytes_for_level_multiplier_additional.size() > 0) {
+      if (FLAGS_max_bytes_for_level_multiplier_additional.size() !=
+          (unsigned int)FLAGS_num_levels) {
+        fprintf(stderr, "Insufficient number of fanouts specified %d\n",
+                (int)FLAGS_max_bytes_for_level_multiplier_additional.size());
+        exit(1);
+      }
+      options.max_bytes_for_level_multiplier_additional =
+        FLAGS_max_bytes_for_level_multiplier_additional;
+    }
     options.level0_stop_writes_trigger = FLAGS_level0_stop_writes_trigger;
     options.level0_file_num_compaction_trigger =
         FLAGS_level0_file_num_compaction_trigger;
@@ -1718,6 +1732,7 @@ int main(int argc, char** argv) {
     long l;
     char junk;
     char hdfsname[2048];
+    char str[512];
     if (leveldb::Slice(argv[i]).starts_with("--benchmarks=")) {
       FLAGS_benchmarks = argv[i] + strlen("--benchmarks=");
     } else if (sscanf(argv[i], "--compression_ratio=%lf%c", &d, &junk) == 1) {
@@ -1853,6 +1868,14 @@ int main(int argc, char** argv) {
     } else if (sscanf(argv[i],"--level0_stop_writes_trigger=%d%c",
         &n, &junk) == 1) {
       FLAGS_level0_stop_writes_trigger = n;
+    } else if (sscanf(argv[i],
+                "--max_bytes_for_level_multiplier_additional=%s%c",
+                str, &junk) == 1) {
+      std::vector<std::string> fanout = leveldb::stringSplit(str, ',');
+      for (unsigned int j= 0; j < fanout.size(); j++) {
+        FLAGS_max_bytes_for_level_multiplier_additional.push_back(
+          std::stoi(fanout[j]));
+      }
     } else if (sscanf(argv[i],"--level0_slowdown_writes_trigger=%d%c",
         &n, &junk) == 1) {
       FLAGS_level0_slowdown_writes_trigger = n;

db/version_set.cc

@@ -1009,7 +1009,8 @@ void VersionSet::Init(int num_levels) {
   for (int i = 0; i < num_levels; i++) {
     if (i > 1) {
       max_file_size_[i] = max_file_size_[i-1] * target_file_size_multiplier;
-      level_max_bytes_[i] = level_max_bytes_[i-1] * max_bytes_multiplier;
+      level_max_bytes_[i] = level_max_bytes_[i-1] * max_bytes_multiplier *
+        options_->max_bytes_for_level_multiplier_additional[i-1];
     } else {
       max_file_size_[i] = options_->target_file_size_base;
       level_max_bytes_[i] = options_->max_bytes_for_level_base;

include/leveldb/options.h

@@ -255,6 +255,12 @@ struct Options {
   // by default 'max_bytes_for_level_base' is 10.
   int max_bytes_for_level_multiplier;
 
+  // Different max-size multipliers for different levels.
+  // These are multiplied by max_bytes_for_level_multiplier to arrive
+  // at the max-size of each level.
+  // Default: 1
+  std::vector<int> max_bytes_for_level_multiplier_additional;
+
   // Maximum number of bytes in all compacted files.  We avoid expanding
   // the lower level file set of a compaction if it would make the
   // total compaction cover more than

util/ldb_cmd.cc

@@ -19,16 +19,6 @@ namespace leveldb {
 
 using namespace std;
 
-vector<string> stringSplit(string arg, char delim) {
-  vector<string> splits;
-  stringstream ss(arg);
-  string item;
-  while(getline(ss, item, delim)) {
-    splits.push_back(item);
-  }
-  return splits;
-}
-
 const string LDBCommand::ARG_DB = "db";
 const string LDBCommand::ARG_HEX = "hex";
 const string LDBCommand::ARG_KEY_HEX = "key_hex";

util/ldb_cmd.h

@@ -19,6 +19,7 @@
 #include "leveldb/slice.h"
 #include "util/logging.h"
 #include "util/ldb_cmd_execute_result.h"
+#include "util/string_util.h"
 #include "utilities/utility_db.h"
 
 using std::string;

util/options.cc

@@ -40,6 +40,7 @@ Options::Options()
       target_file_size_multiplier(1),
       max_bytes_for_level_base(10 * 1048576),
       max_bytes_for_level_multiplier(10),
+      max_bytes_for_level_multiplier_additional(num_levels, 1),
       expanded_compaction_factor(25),
       source_compaction_factor(1),
       max_grandparent_overlap_factor(10),
@@ -142,6 +143,10 @@ Options::Dump(Logger* log) const
         max_bytes_for_level_base);
     Log(log,"         Options.max_bytes_for_level_multiplier: %d",
         max_bytes_for_level_multiplier);
+    for (int i = 0; i < num_levels; i++) {
+      Log(log,"Options.max_bytes_for_level_multiplier_addtl[%d]: %d",
+          i, max_bytes_for_level_multiplier_additional[i]);
+    }
     Log(log,"             Options.expanded_compaction_factor: %d",
         expanded_compaction_factor);
     Log(log,"               Options.source_compaction_factor: %d",

util/string_util.cc

@@ -0,0 +1,23 @@
+// Copyright (c) 2013 Facebook.
+
+#include <sstream>
+#include <string>
+#include <vector>
+
+namespace leveldb {
+
+using namespace std;
+using std::string;
+using std::vector;
+using std::stringstream;
+
+vector<string> stringSplit(string arg, char delim) {
+  vector<string> splits;
+  stringstream ss(arg);
+  string item;
+  while(getline(ss, item, delim)) {
+    splits.push_back(item);
+  }
+  return splits;
+}
+}

util/string_util.h

@@ -0,0 +1,12 @@
+// Copyright (c) 2013 Facebook.
+
+#ifndef STORAGE_LEVELDB_UTIL_STRINGUTIL_H_
+#define STORAGE_LEVELDB_UTIL_STRINGUTIL_H_
+
+namespace leveldb {
+
+extern std::vector<std::string> stringSplit(std::string arg, char delim);
+
+}
+
+#endif  // STORAGE_LEVELDB_UTIL_STRINGUTIL_H_