feat(server): WIP memory fragmentation WIP, high level flow only drag…

…onflydb#448

Signed-off-by: Boaz Sade <boaz@dragonflydb.io>

src/server/engine_shard_set.cc

@@ -11,10 +11,12 @@ extern "C" {
 
 #include "base/flags.h"
 #include "base/logging.h"
+#include "base/proc_util.h"
 #include "server/blocking_controller.h"
 #include "server/server_state.h"
 #include "server/tiered_storage.h"
 #include "server/transaction.h"
+#include "strings/human_readable.h"
 #include "util/fiber_sched_algo.h"
 #include "util/varz.h"
 
@@ -30,6 +32,17 @@ ABSL_FLAG(bool, cache_mode, false,
           "If true, the backend behaves like a cache, "
           "by evicting entries when getting close to maxmemory limit");
 
+// memory de-fragmentation related flags
+ABSL_FLAG(uint64_t, mem_defrag_threshold, 20,
+          "Threshold level to run memory fragmentation between total available memory and "
+          "currently commited memory");
+
+ABSL_FLAG(float, commit_use_threshold, 1.3,
+          "The threshold to apply memory fragmentation between commited and used memory");
+
+ABSL_FLAG(float, mem_utilization_threshold, 0.5,
+          "Threshold for memory page under utilization. Bellow this threshold a pointer in this "
+          "page will be moved in memory");
 namespace dfly {
 
 using namespace util;
@@ -39,8 +52,45 @@ using absl::GetFlag;
 
 namespace {
 
+constexpr DbIndex DEFAULT_DB_INDEX = 0;
+
 vector<EngineShardSet::CachedStats> cached_stats;  // initialized in EngineShardSet::Init
 
+inline bool MemUsageChanged(uint64_t current, uint64_t now) {
+  static const float commit_use_threshold = GetFlag(FLAGS_commit_use_threshold);
+  // we only care whether the memory usage is going up
+  // otherwise we don't really case
+  return (current * commit_use_threshold) < float(now);
+}
+
+// This function checks 3 things:
+// 1. Don't try memory fragmentation if we don't use "enough" memory (control by
+// mem_defrag_threshold flag)
+// 2. That we had change in memory usage - to prevent endless loop
+// 3. in case the above is OK, make sure that we have a "gap" between usage and commited memory
+// (control by commit_use_threshold flag)
+bool IsMemDefragRequired(uint64_t* current_commit, uint64_t* current_use) {
+  static const uint64_t mem_size = max_memory_limit;
+  static const uint64_t threshold_mem = mem_size / GetFlag(FLAGS_mem_defrag_threshold);
+  static const float commit_use_threshold = GetFlag(FLAGS_commit_use_threshold);
+
+  uint64_t commited = GetMallocCurrentCommitted();
+  uint64_t mem_in_use = used_mem_current.load(memory_order_relaxed);
+
+  // we want to make sure that we are not running this to many times - i.e.
+  // if there was no change to the memory, don't run this
+  if (MemUsageChanged(*current_commit, commited) || MemUsageChanged(*current_use, mem_in_use)) {
+    if (threshold_mem < commited && mem_in_use != 0 &&
+        (float(mem_in_use * commit_use_threshold) < commited)) {
+      // we have way more commited then actual usage
+      *current_commit = commited;
+      *current_use = mem_in_use;
+      return true;
+    }
+  }
+  return false;
+}
+
 }  // namespace
 
 constexpr size_t kQueueLen = 64;
@@ -56,6 +106,41 @@ EngineShard::Stats& EngineShard::Stats::operator+=(const EngineShard::Stats& o)
   return *this;
 }
 
+// for now this does nothing
+uint64_t EngineShard::DoDefragIfRequired(uint64_t cursor) {
+  // Placeholder for now!
+  return cursor;
+}
+
+// the memory de-fragmenting task is as follow:
+//  1. Check if memory usage is high enough
+//  2. Check if diff between commited and used memory is high enough
+//  3. Check if we have memory changes (to ensure that we not running endlessly).
+//  4. if all the above pass -> run on the shard and try to de-fragment memory by re-allocating
+//  values
+//     if the cursor for this is signal that we are not done, schedule the task to run at high
+//     priority otherwise lower the task priority so that it would not use the CPU when not required
+void EngineShard::StartDefragTask(util::ProactorBase* pb) {
+  uint64_t commited = GetMallocCurrentCommitted();
+  uint64_t mem_in_use = used_mem_current.load(memory_order_relaxed);
+
+  defrag_task_ = pb->AddOnIdleTask([this, commited, mem_in_use, cursor = 0ul]() mutable {
+    const auto shard_id = db_slice().shard_id();
+    bool required_state = IsMemDefragRequired(&commited, &mem_in_use);
+    if (required_state) {
+      VLOG(1) << shard_id << ": need to update memory - commited memory "
+              << strings::HumanReadableNumBytes(commited) << ", the use memory is "
+              << strings::HumanReadableNumBytes(mem_in_use);
+      cursor = DoDefragIfRequired(cursor);  // we are only checking for DB == 0
+      if (cursor > 0) {
+        return util::ProactorBase::kOnIdleMaxLevel;  // we need to run more of this
+      }
+    }
+    // by default we just want to not get in the way..
+    return 0u;
+  });
+}
+
 EngineShard::EngineShard(util::ProactorBase* pb, bool update_db_time, mi_heap_t* heap)
     : queue_(kQueueLen), txq_([](const Transaction* t) { return t->txid(); }), mi_resource_(heap),
       db_slice_(pb->GetIndex(), GetFlag(FLAGS_cache_mode), this) {
@@ -75,6 +160,8 @@ EngineShard::EngineShard(util::ProactorBase* pb, bool update_db_time, mi_heap_t*
   tmp_str1 = sdsempty();
 
   db_slice_.UpdateExpireBase(absl::GetCurrentTimeNanos() / 1000000, 0);
+
+  StartDefragTask(pb);
 }
 
 EngineShard::~EngineShard() {
@@ -92,6 +179,8 @@ void EngineShard::Shutdown() {
   if (periodic_task_) {
     ProactorBase::me()->CancelPeriodic(periodic_task_);
   }
+
+  ProactorBase::me()->RemoveOnIdleTask(defrag_task_);
 }
 
 void EngineShard::InitThreadLocal(ProactorBase* pb, bool update_db_time) {

src/server/engine_shard_set.h

@@ -153,6 +153,16 @@ class EngineShard {
 
   void CacheStats();
 
+  // We are running a task that checks whether we need to
+  // do memory de-fragmentation here, this task only run
+  // when there are available CPU time.
+  void StartDefragTask(util::ProactorBase* pb);
+
+  // scan the shard with the cursor and apply
+  // de-fragmentation option for entries. This function will return the new cursor at the end of the
+  // scan This function is called from context of StartDefragTask
+  uint64_t DoDefragIfRequired(uint64_t cursor);
+
   ::util::fibers_ext::FiberQueue queue_;
   ::boost::fibers::fiber fiber_q_;
 
@@ -170,6 +180,7 @@ class EngineShard {
   IntentLock shard_lock_;
 
   uint32_t periodic_task_ = 0;
+  uint32_t defrag_task_ = 0;
   std::unique_ptr<TieredStorage> tiered_storage_;
   std::unique_ptr<BlockingController> blocking_controller_;