diff --git a/python_module/megengine/data/_queue.py b/python_module/megengine/data/_queue.py
index f90f090dd59b3114284ba9e50bad3db95a4fde23..94898e99ccc0784ce8f93f457dfb7472f3b90e72 100644
--- a/python_module/megengine/data/_queue.py
+++ b/python_module/megengine/data/_queue.py
@@ -16,6 +16,19 @@ import pyarrow.plasma as plasma
 
 MGE_PLASMA_MEMORY = int(os.environ.get("MGE_PLASMA_MEMORY", 4000000000))  # 4GB
 
+# Each process only need to start one plasma store, so we set it as a global variable.
+# TODO: how to share between different processes?
+MGE_PLASMA_STORE_MANAGER = None
+
+
+def _clear_plasma_store():
+    # `_PlasmaStoreManager.__del__` will not ne called automaticly in subprocess,
+    # so this function should be called explicitly
+    global MGE_PLASMA_STORE_MANAGER
+    if MGE_PLASMA_STORE_MANAGER is not None:
+        del MGE_PLASMA_STORE_MANAGER
+        MGE_PLASMA_STORE_MANAGER = None
+
 
 class _PlasmaStoreManager:
     def __init__(self):
@@ -34,11 +47,6 @@ class _PlasmaStoreManager:
             self.plasma_store.kill()
 
 
-# Each process only need to start one plasma store, so we set it as a global variable.
-# TODO: how to share between different processes?
-MGE_PLASMA_STORE_MANAGER = _PlasmaStoreManager()
-
-
 class PlasmaShmQueue:
     def __init__(self, maxsize: int = 0):
         r"""Use pyarrow in-memory plasma store to implement shared memory queue.
@@ -51,6 +59,17 @@ class PlasmaShmQueue:
         :param maxsize: maximum size of the queue, `None` means no limit. (default: ``None``)
         """
 
+        # Lazy start the plasma store manager
+        global MGE_PLASMA_STORE_MANAGER
+        if MGE_PLASMA_STORE_MANAGER is None:
+            try:
+                MGE_PLASMA_STORE_MANAGER = _PlasmaStoreManager()
+            except FileNotFoundError as e:
+                raise FileNotFoundError(
+                    "command 'plasma_store' not found in your $PATH!"
+                    "Please make sure pyarrow installed and add into $PATH."
+                )
+
         self.socket_name = MGE_PLASMA_STORE_MANAGER.socket_name
 
         # TODO: how to catch the exception happened in `plasma.connect`?
@@ -100,6 +119,7 @@ class PlasmaShmQueue:
     def close(self):
         self.queue.close()
         self.disconnect_client()
+        _clear_plasma_store()
 
     def cancel_join_thread(self):
         self.queue.cancel_join_thread()
diff --git a/python_module/megengine/data/dataloader.py b/python_module/megengine/data/dataloader.py
index d6e03c8bb3e0f56c3f7f01b1fedc9a596df23005..388a8f0e657a9bc03364b2ecc36f439b53d220bc 100644
--- a/python_module/megengine/data/dataloader.py
+++ b/python_module/megengine/data/dataloader.py
@@ -17,12 +17,13 @@ import numpy as np
 
 import megengine as mge
 
+from ..logger import get_logger
 from .collator import Collator
 from .dataset import Dataset
 from .sampler import Sampler, SequentialSampler
 from .transform import PseudoTransform, Transform
 
-logger = mge.get_logger(__name__)
+logger = get_logger(__name__)
 
 
 MP_QUEUE_GET_TIMEOUT = 5
@@ -167,7 +168,7 @@ class _SerialDataLoaderIter(_BaseDataLoaderIter):
 
 
 class _ParallelDataLoaderIter(_BaseDataLoaderIter):
-    __initialzed = False
+    __initialized = False
 
     def __init__(self, loader):
         super(_ParallelDataLoaderIter, self).__init__(loader)