diff --git a/core/src/cache/CacheMgr.h b/core/src/cache/CacheMgr.h
index c7ab9e26ab..114b089c8f 100644
--- a/core/src/cache/CacheMgr.h
+++ b/core/src/cache/CacheMgr.h
@@ -57,12 +57,6 @@ class CacheMgr {
     void
     SetCapacity(int64_t capacity);
 
-    void
-    Lock();
-
-    void
-    Unlock();
-
  protected:
     CacheMgr();
 
@@ -70,7 +64,6 @@ class CacheMgr {
 
  protected:
     std::shared_ptr<Cache<ItemObj>> cache_;
-    std::mutex mutex_;
 };
 
 }  // namespace cache
diff --git a/core/src/cache/CacheMgr.inl b/core/src/cache/CacheMgr.inl
index 606510c66c..b45b403595 100644
--- a/core/src/cache/CacheMgr.inl
+++ b/core/src/cache/CacheMgr.inl
@@ -133,17 +133,5 @@ CacheMgr<ItemObj>::SetCapacity(int64_t capacity) {
     cache_->set_capacity(capacity);
 }
 
-template <typename ItemObj>
-void
-CacheMgr<ItemObj>::Lock() {
-    mutex_.lock();
-}
-
-template <typename ItemObj>
-void
-CacheMgr<ItemObj>::Unlock() {
-    mutex_.unlock();
-}
-
 }  // namespace cache
 }  // namespace milvus
diff --git a/core/src/cache/GpuCacheMgr.cpp b/core/src/cache/GpuCacheMgr.cpp
index 30e4ec4c5a..b371f8f57a 100644
--- a/core/src/cache/GpuCacheMgr.cpp
+++ b/core/src/cache/GpuCacheMgr.cpp
@@ -84,9 +84,7 @@ GpuCacheMgr::GetInstance(int64_t gpu_id) {
 void
 GpuCacheMgr::OnGpuCacheCapacityChanged(int64_t capacity) {
     for (auto& iter : instance_) {
-        iter.second->Lock();
         iter.second->SetCapacity(capacity * G_BYTE);
-        iter.second->Unlock();
     }
 }
 
diff --git a/core/src/db/engine/ExecutionEngineImpl.cpp b/core/src/db/engine/ExecutionEngineImpl.cpp
index e367a35907..8be6ec8898 100644
--- a/core/src/db/engine/ExecutionEngineImpl.cpp
+++ b/core/src/db/engine/ExecutionEngineImpl.cpp
@@ -555,11 +555,9 @@ ExecutionEngineImpl::CopyToGpu(uint64_t device_id, bool hybrid) {
              */
             ENGINE_LOG_DEBUG << "CPU to GPU" << device_id << " start";
             auto gpu_cache_mgr = cache::GpuCacheMgr::GetInstance(device_id);
-            // gpu_cache_mgr->Lock();
             // gpu_cache_mgr->Reserve(index_->Size());
             index_ = knowhere::cloner::CopyCpuToGpu(index_, device_id, knowhere::Config());
             // gpu_cache_mgr->InsertItem(location_, std::static_pointer_cast<cache::DataObj>(index_));
-            // gpu_cache_mgr->Unlock();
             ENGINE_LOG_DEBUG << "CPU to GPU" << device_id << " finished";
         } catch (std::exception& e) {
             ENGINE_LOG_ERROR << e.what();
@@ -577,10 +575,8 @@ ExecutionEngineImpl::CopyToIndexFileToGpu(uint64_t device_id) {
     // the ToIndexData is only a placeholder, cpu-copy-to-gpu action is performed in
     if (index_) {
         auto gpu_cache_mgr = milvus::cache::GpuCacheMgr::GetInstance(device_id);
-        gpu_cache_mgr->Lock();
         gpu_num_ = device_id;
         gpu_cache_mgr->Reserve(index_->Size());
-        gpu_cache_mgr->Unlock();
     }
 #endif
     return Status::OK();
@@ -964,10 +960,8 @@ ExecutionEngineImpl::GetVectorByID(const int64_t& id, uint8_t* vector, bool hybr
 Status
 ExecutionEngineImpl::Cache() {
     auto cpu_cache_mgr = milvus::cache::CpuCacheMgr::GetInstance();
-    cpu_cache_mgr->Lock();
     cache::DataObjPtr obj = std::static_pointer_cast<cache::DataObj>(index_);
     cpu_cache_mgr->InsertItem(location_, obj);
-    cpu_cache_mgr->Unlock();
     return Status::OK();
 }