diff --git a/data/database.py b/data/database.py
index ccfad8b82..cad01b4ac 100644
--- a/data/database.py
+++ b/data/database.py
@@ -11,6 +11,7 @@ from random import SystemRandom
 from datetime import datetime
 from peewee import *
 from data.read_slave import ReadSlaveModel
+from data.fields import ResumableSHAField, JSONField
 from sqlalchemy.engine.url import make_url
 from collections import defaultdict
 
@@ -750,35 +751,13 @@ class RepositoryAuthorizedEmail(BaseModel):
     )
 
 
-class ResumableSHAField(TextField):
-  def db_value(self, value):
-    sha_state = value.state()
-
-    # One of the fields is a byte string, let's base64 encode it to make sure
-    # we can store and fetch it regardless of default collocation
-    sha_state[3] = base64.b64encode(sha_state[3])
-
-    return json.dumps(sha_state)
-
-  def python_value(self, value):
-    to_resume = resumablehashlib.sha256()
-    if value is None:
-      return to_resume
-
-    sha_state = json.loads(value)
-
-    # We need to base64 decode the data bytestring
-    sha_state[3] = base64.b64decode(sha_state[3])
-    to_resume.set_state(sha_state)
-    return to_resume
-
-
 class BlobUpload(BaseModel):
   repository = ForeignKeyField(Repository, index=True)
   uuid = CharField(index=True, unique=True)
   byte_count = IntegerField(default=0)
   sha_state = ResumableSHAField(null=True, default=resumablehashlib.sha256)
   location = ForeignKeyField(ImageStorageLocation)
+  storage_metadata = JSONField(null=True, default={})
 
   class Meta:
     database = db
diff --git a/data/fields.py b/data/fields.py
new file mode 100644
index 000000000..123811ccd
--- /dev/null
+++ b/data/fields.py
@@ -0,0 +1,38 @@
+import base64
+import resumablehashlib
+import json
+
+from peewee import TextField
+
+
+class ResumableSHAField(TextField):
+  def db_value(self, value):
+    sha_state = value.state()
+
+    # One of the fields is a byte string, let's base64 encode it to make sure
+    # we can store and fetch it regardless of default collocation.
+    sha_state[3] = base64.b64encode(sha_state[3])
+
+    return json.dumps(sha_state)
+
+  def python_value(self, value):
+    to_resume = resumablehashlib.sha256()
+    if value is None:
+      return to_resume
+
+    sha_state = json.loads(value)
+
+    # We need to base64 decode the data bytestring.
+    sha_state[3] = base64.b64decode(sha_state[3])
+    to_resume.set_state(sha_state)
+    return to_resume
+
+
+class JSONField(TextField):
+  def db_value(self, value):
+    return json.dumps(value)
+
+  def python_value(self, value):
+    if value is None or value == "":
+      return {}
+    return json.loads(value)
diff --git a/data/model/blob.py b/data/model/blob.py
index 5820ba3b1..a547a6f97 100644
--- a/data/model/blob.py
+++ b/data/model/blob.py
@@ -67,7 +67,8 @@ def get_blob_upload(namespace, repo_name, upload_uuid):
     raise InvalidBlobUpload()
 
 
-def initiate_upload(namespace, repo_name, uuid, location_name):
+def initiate_upload(namespace, repo_name, uuid, location_name, storage_metadata):
   repo = _basequery.get_existing_repository(namespace, repo_name)
   location = ImageStorageLocation.get(name=location_name)
-  return BlobUpload.create(repository=repo, location=location, uuid=uuid)
+  return BlobUpload.create(repository=repo, location=location, uuid=uuid,
+                           storage_metadata=storage_metadata)
diff --git a/endpoints/v2/blob.py b/endpoints/v2/blob.py
index 705794b6b..9867917a1 100644
--- a/endpoints/v2/blob.py
+++ b/endpoints/v2/blob.py
@@ -119,8 +119,8 @@ def _render_range(num_uploaded_bytes, with_bytes_prefix=True):
 @anon_protect
 def start_blob_upload(namespace, repo_name):
   location_name = storage.preferred_locations[0]
-  new_upload_uuid = storage.initiate_chunked_upload(location_name)
-  model.blob.initiate_upload(namespace, repo_name, new_upload_uuid, location_name)
+  new_upload_uuid, upload_metadata = storage.initiate_chunked_upload(location_name)
+  model.blob.initiate_upload(namespace, repo_name, new_upload_uuid, location_name, upload_metadata)
 
   digest = request.args.get('digest', None)
   if digest is None:
@@ -205,11 +205,13 @@ def _upload_chunk(namespace, repo_name, upload_uuid):
   input_fp = wrap_with_handler(input_fp, found.sha_state.update)
 
   try:
-    length_written = storage.stream_upload_chunk({found.location.name}, upload_uuid, start_offset,
-                                                 length, input_fp)
+    length_written, new_metadata = storage.stream_upload_chunk({found.location.name}, upload_uuid,
+                                                               start_offset, length, input_fp,
+                                                               found.storage_metadata)
   except InvalidChunkException:
     _range_not_satisfiable(found.byte_count)
 
+  found.storage_metadata = new_metadata
   found.byte_count += length_written
   return found
 
@@ -222,7 +224,8 @@ def _finish_upload(namespace, repo_name, upload_obj, expected_digest):
 
   # Mark the blob as uploaded.
   final_blob_location = digest_tools.content_path(expected_digest)
-  storage.complete_chunked_upload({upload_obj.location.name}, upload_obj.uuid, final_blob_location)
+  storage.complete_chunked_upload({upload_obj.location.name}, upload_obj.uuid, final_blob_location,
+                                  upload_obj.storage_metadata)
   model.blob.store_blob_record_and_temp_link(namespace, repo_name, expected_digest,
                                              upload_obj.location, upload_obj.byte_count,
                                              app.config['PUSH_TEMP_TAG_EXPIRATION_SEC'])
@@ -278,6 +281,6 @@ def cancel_upload(namespace, repo_name, upload_uuid):
   # We delete the record for the upload first, since if the partial upload in
   # storage fails to delete, it doesn't break anything
   found.delete_instance()
-  storage.cancel_chunked_upload({found.location.name}, found.uuid)
+  storage.cancel_chunked_upload({found.location.name}, found.uuid, found.storage_metadata)
 
   return make_response('', 204)
diff --git a/storage/basestorage.py b/storage/basestorage.py
index 55035901c..af51a45e3 100644
--- a/storage/basestorage.py
+++ b/storage/basestorage.py
@@ -42,18 +42,9 @@ class StoragePaths(object):
 
 
 class BaseStorage(StoragePaths):
-  """Storage is organized as follow:
-  $ROOT/images/<image_id>/json
-  $ROOT/images/<image_id>/layer
-  $ROOT/repositories/<namespace>/<repository_name>/<tag_name>
-  """
-
-  # Useful if we want to change those locations later without rewriting
-  # the code which uses Storage
-  repositories = 'repositories'
-  images = 'images'
-  # Set the IO buffer to 64kB
-  buffer_size = 64 * 1024
+  def __init__(self):
+    # Set the IO buffer to 64kB
+    self.buffer_size = 64 * 1024
 
   def setup(self):
     """ Called to perform any storage system setup. """
@@ -99,31 +90,55 @@ class BaseStorage(StoragePaths):
   def get_checksum(self, path):
     raise NotImplementedError
 
+  def stream_write_to_fp(self, in_fp, out_fp, num_bytes=-1):
+    """ Copy the specified number of bytes from the input file stream to the output stream. If
+        num_bytes < 0 copy until the stream ends.
+    """
+    bytes_copied = 0
+    while bytes_copied < num_bytes or num_bytes == -1:
+      size_to_read = min(num_bytes - bytes_copied, self.buffer_size)
+      if size_to_read < 0:
+        size_to_read = self.buffer_size
+
+      try:
+        buf = in_fp.read(size_to_read)
+        if not buf:
+          break
+        out_fp.write(buf)
+        bytes_copied += len(buf)
+      except IOError:
+        break
+
+    return bytes_copied
+
 
 class InvalidChunkException(RuntimeError):
   pass
 
 
 class BaseStorageV2(BaseStorage):
-  def initiate_chunked_upload(self, upload_uuid):
-    """ Start a new chunked upload
+  def initiate_chunked_upload(self):
+    """ Start a new chunked upload, returning the uuid and any associated storage metadata
     """
     raise NotImplementedError
 
-  def stream_upload_chunk(self, uuid, offset, length, in_fp, hash_obj):
+  def stream_upload_chunk(self, uuid, offset, length, in_fp, storage_metadata):
     """ Upload the specified amount of data from the given file pointer to the chunked destination
-        specified, starting at the given offset. Raises InvalidChunkException if the offset or
-        length can not be accepted.
+        specified, starting at the given offset. Returns the number of bytes uploaded, and a new
+        version of the storage_metadata. Raises InvalidChunkException if the offset or length can
+        not be accepted.
     """
     raise NotImplementedError
 
-  def complete_chunked_upload(self, uuid, final_path):
+  def complete_chunked_upload(self, uuid, final_path, storage_metadata):
     """ Complete the chunked upload and store the final results in the path indicated.
+        Returns nothing.
     """
     raise NotImplementedError
 
-  def cancel_chunked_upload(self, uuid):
+  def cancel_chunked_upload(self, uuid, storage_metadata):
     """ Cancel the chunked upload and clean up any outstanding partially uploaded data.
+        Returns nothing.
     """
     raise NotImplementedError
 
diff --git a/storage/cloud.py b/storage/cloud.py
index 91dadfb3e..91d2b3fdc 100644
--- a/storage/cloud.py
+++ b/storage/cloud.py
@@ -3,18 +3,25 @@ import os
 import logging
 
 import boto.s3.connection
+import boto.s3.multipart
 import boto.gs.connection
 import boto.s3.key
 import boto.gs.key
 
 from io import BufferedIOBase
+from uuid import uuid4
 
-from storage.basestorage import BaseStorage
+from storage.basestorage import BaseStorageV2, InvalidChunkException
 
 
 logger = logging.getLogger(__name__)
 
 
+_MULTIPART_UPLOAD_ID_KEY = 'upload_id'
+_LAST_PART_KEY = 'last_part_num'
+_LAST_CHUNK_ENCOUNTERED = 'last_chunk_encountered'
+
+
 class StreamReadKeyAsFile(BufferedIOBase):
   def __init__(self, key):
     self._key = key
@@ -37,9 +44,13 @@ class StreamReadKeyAsFile(BufferedIOBase):
     self._key.close(fast=True)
 
 
-class _CloudStorage(BaseStorage):
+class _CloudStorage(BaseStorageV2):
   def __init__(self, connection_class, key_class, connect_kwargs, upload_params, storage_path,
                access_key, secret_key, bucket_name):
+    super(_CloudStorage, self).__init__()
+
+    self.upload_chunk_size = 5 * 1024 * 1024
+
     self._initialized = False
     self._bucket_name = bucket_name
     self._access_key = access_key
@@ -135,12 +146,9 @@ class _CloudStorage(BaseStorage):
       raise IOError('No such key: \'{0}\''.format(path))
     return StreamReadKeyAsFile(key)
 
-  def stream_write(self, path, fp, content_type=None, content_encoding=None):
+  def __initiate_multipart_upload(self, path, content_type, content_encoding):
     # Minimum size of upload part size on S3 is 5MB
     self._initialize_cloud_conn()
-    buffer_size = 5 * 1024 * 1024
-    if self.buffer_size > buffer_size:
-      buffer_size = self.buffer_size
     path = self._init_path(path)
 
     metadata = {}
@@ -150,16 +158,20 @@ class _CloudStorage(BaseStorage):
     if content_encoding is not None:
       metadata['Content-Encoding'] = content_encoding
 
-    mp = self._cloud_bucket.initiate_multipart_upload(path, metadata=metadata,
-                                                      **self._upload_params)
+    return self._cloud_bucket.initiate_multipart_upload(path, metadata=metadata,
+                                                        **self._upload_params)
+
+  def stream_write(self, path, fp, content_type=None, content_encoding=None):
+    mp = self.__initiate_multipart_upload(path, content_type, content_encoding)
     num_part = 1
     while True:
       try:
-        buf = fp.read(buffer_size)
-        if not buf:
+        buf = StringIO.StringIO()
+        bytes_written = self.stream_write_to_fp(fp, buf, self.upload_chunk_size)
+        if bytes_written == 0:
           break
-        io = StringIO.StringIO(buf)
-        mp.upload_part_from_file(io, num_part)
+
+        mp.upload_part_from_file(buf, num_part)
         num_part += 1
         io.close()
       except IOError:
@@ -217,6 +229,57 @@ class _CloudStorage(BaseStorage):
 
     return k.etag[1:-1][:7]
 
+  def _rel_upload_path(self, uuid):
+    return 'uploads/{0}'.format(uuid)
+
+  def initiate_chunked_upload(self):
+    self._initialize_cloud_conn()
+    random_uuid = str(uuid4())
+    path = self._init_path(self._rel_upload_path(random_uuid))
+    mpu = self.__initiate_multipart_upload(path, content_type=None, content_encoding=None)
+
+    metadata = {
+      _MULTIPART_UPLOAD_ID_KEY: mpu.id,
+      _LAST_PART_KEY: 0,
+      _LAST_CHUNK_ENCOUNTERED: False,
+    }
+
+    return mpu.id, metadata
+
+  def _get_multipart_upload_key(self, uuid, storage_metadata):
+    mpu = boto.s3.multipart.MultiPartUpload(self._cloud_bucket)
+    mpu.id = storage_metadata[_MULTIPART_UPLOAD_ID_KEY]
+    mpu.key = self._init_path(self._rel_upload_path(uuid))
+    return mpu
+
+  def stream_upload_chunk(self, uuid, offset, length, in_fp, storage_metadata):
+    self._initialize_cloud_conn()
+    mpu = self._get_multipart_upload_key(uuid, storage_metadata)
+    last_part_num = storage_metadata[_LAST_PART_KEY]
+
+    if storage_metadata[_LAST_CHUNK_ENCOUNTERED] and length != 0:
+      msg = 'Length must be at least the the upload chunk size: %s' % self.upload_chunk_size
+      raise InvalidChunkException(msg)
+
+    part_num = last_part_num + 1
+    mpu.upload_part_from_file(in_fp, part_num, length)
+
+    new_metadata = {
+      _MULTIPART_UPLOAD_ID_KEY: mpu.id,
+      _LAST_PART_KEY: part_num,
+      _LAST_CHUNK_ENCOUNTERED: True,
+    }
+
+    return length, new_metadata
+
+  def complete_chunked_upload(self, uuid, final_path, storage_metadata):
+    mpu = self._get_multipart_upload_key(uuid, storage_metadata)
+    mpu.complete_upload()
+
+  def cancel_chunked_upload(self, uuid, storage_metadata):
+    mpu = self._get_multipart_upload_key(uuid, storage_metadata)
+    mpu.cancel_multipart_upload()
+
 
 class S3Storage(_CloudStorage):
   def __init__(self, storage_path, s3_access_key, s3_secret_key, s3_bucket):
diff --git a/storage/fakestorage.py b/storage/fakestorage.py
index f351ca150..d72b5ddc4 100644
--- a/storage/fakestorage.py
+++ b/storage/fakestorage.py
@@ -1,8 +1,14 @@
-from storage.basestorage import BaseStorage
+import cStringIO as StringIO
+import hashlib
 
-_FAKE_STORAGE_MAP = {}
+from collections import defaultdict
+from uuid import uuid4
 
-class FakeStorage(BaseStorage):
+from storage.basestorage import BaseStorageV2
+
+_FAKE_STORAGE_MAP = defaultdict(StringIO.StringIO)
+
+class FakeStorage(BaseStorageV2):
   def _init_path(self, path=None, create=False):
     return path
 
@@ -10,16 +16,26 @@ class FakeStorage(BaseStorage):
     if not path in _FAKE_STORAGE_MAP:
       raise IOError('Fake file %s not found' % path)
 
-    return _FAKE_STORAGE_MAP.get(path)
+    _FAKE_STORAGE_MAP.get(path).seek(0)
+    return _FAKE_STORAGE_MAP.get(path).read()
 
   def put_content(self, path, content):
-    _FAKE_STORAGE_MAP[path] = content
+    _FAKE_STORAGE_MAP.pop(path, None)
+    _FAKE_STORAGE_MAP[path].write(content)
 
   def stream_read(self, path):
-    yield _FAKE_STORAGE_MAP[path]
+    io_obj = _FAKE_STORAGE_MAP[path]
+    io_obj.seek(0)
+    while True:
+      buf = io_obj.read(self.buffer_size)
+      if not buf:
+        break
+      yield buf
 
   def stream_write(self, path, fp, content_type=None, content_encoding=None):
-    _FAKE_STORAGE_MAP[path] = fp.read()
+    out_fp = _FAKE_STORAGE_MAP[path]
+    out_fp.seek(0)
+    self.stream_write_to_fp(fp, out_fp)
 
   def remove(self, path):
     _FAKE_STORAGE_MAP.pop(path, None)
@@ -28,4 +44,21 @@ class FakeStorage(BaseStorage):
     return path in _FAKE_STORAGE_MAP
 
   def get_checksum(self, path):
-    return path
+    return hashlib.sha256(_FAKE_STORAGE_MAP[path].read()).hexdigest()[:7]
+
+  def initiate_chunked_upload(self):
+    new_uuid = str(uuid4())
+    _FAKE_STORAGE_MAP[new_uuid].seek(0)
+    return new_uuid, {}
+
+  def stream_upload_chunk(self, uuid, offset, length, in_fp, _):
+    upload_storage = _FAKE_STORAGE_MAP[uuid]
+    upload_storage.seek(offset)
+    return self.stream_write_to_fp(in_fp, upload_storage, length)
+
+  def complete_chunked_upload(self, uuid, final_path, _):
+    _FAKE_STORAGE_MAP[final_path] = _FAKE_STORAGE_MAP[uuid]
+    _FAKE_STORAGE_MAP.pop(uuid, None)
+
+  def cancel_chunked_upload(self, uuid, _):
+    _FAKE_STORAGE_MAP.pop(uuid, None)
diff --git a/storage/local.py b/storage/local.py
index 9495aed9c..1753e4d85 100644
--- a/storage/local.py
+++ b/storage/local.py
@@ -14,8 +14,8 @@ logger = logging.getLogger(__name__)
 
 
 class LocalStorage(BaseStorageV2):
-
   def __init__(self, storage_path):
+    super(LocalStorage, self).__init__()
     self._root_path = storage_path
 
   def _init_path(self, path=None, create=False):
@@ -54,28 +54,7 @@ class LocalStorage(BaseStorageV2):
     # Size is mandatory
     path = self._init_path(path, create=True)
     with open(path, mode='wb') as out_fp:
-      self._stream_write_to_fp(fp, out_fp)
-
-  def _stream_write_to_fp(self, in_fp, out_fp, num_bytes=-1):
-    """ Copy the specified number of bytes from the input file stream to the output stream. If
-        num_bytes < 0 copy until the stream ends.
-    """
-    bytes_copied = 0
-    while bytes_copied < num_bytes or num_bytes == -1:
-      size_to_read = min(num_bytes - bytes_copied, self.buffer_size)
-      if size_to_read < 0:
-        size_to_read = self.buffer_size
-
-      try:
-        buf = in_fp.read(size_to_read)
-        if not buf:
-          break
-        out_fp.write(buf)
-        bytes_copied += len(buf)
-      except IOError:
-        break
-
-    return bytes_copied
+      self.stream_write_to_fp(fp, out_fp)
 
   def list_directory(self, path=None):
     path = self._init_path(path)
@@ -124,14 +103,14 @@ class LocalStorage(BaseStorageV2):
     with open(self._init_path(self._rel_upload_path(new_uuid), create=True), 'w'):
       pass
 
-    return new_uuid
+    return new_uuid, {}
 
-  def stream_upload_chunk(self, uuid, offset, length, in_fp):
+  def stream_upload_chunk(self, uuid, offset, length, in_fp, _):
     with open(self._init_path(self._rel_upload_path(uuid)), 'r+b') as upload_storage:
       upload_storage.seek(offset)
-      return self._stream_write_to_fp(in_fp, upload_storage, length)
+      return self.stream_write_to_fp(in_fp, upload_storage, length), {}
 
-  def complete_chunked_upload(self, uuid, final_path):
+  def complete_chunked_upload(self, uuid, final_path, _):
     content_path = self._rel_upload_path(uuid)
     final_path_abs = self._init_path(final_path, create=True)
     if not self.exists(final_path_abs):
@@ -140,7 +119,7 @@ class LocalStorage(BaseStorageV2):
     else:
       logger.debug('Content already exists at path: %s', final_path_abs)
 
-  def cancel_chunked_upload(self, uuid):
+  def cancel_chunked_upload(self, uuid, _):
     content_path = self._init_path(self._rel_upload_path(uuid))
     os.remove(content_path)
 
diff --git a/storage/swift.py b/storage/swift.py
index 73f743290..a3e3bcfb0 100644
--- a/storage/swift.py
+++ b/storage/swift.py
@@ -13,6 +13,8 @@ logger = logging.getLogger(__name__)
 class SwiftStorage(BaseStorage):
   def __init__(self, swift_container, storage_path, auth_url, swift_user,
                swift_password, auth_version=None, os_options=None, ca_cert_path=None):
+    super(SwiftStorage, self).__init__()
+
     self._swift_container = swift_container
     self._storage_path = storage_path
 
diff --git a/test/data/test.db b/test/data/test.db
index e0423a2c8..4bc9e762c 100644
Binary files a/test/data/test.db and b/test/data/test.db differ