diff --git a/gns3server/controller/export_project.py b/gns3server/controller/export_project.py
index 9afcd906..c2641364 100644
--- a/gns3server/controller/export_project.py
+++ b/gns3server/controller/export_project.py
@@ -22,7 +22,6 @@ import asyncio
 import aiohttp
 import zipfile
 import tempfile
-import zipstream
 
 from datetime import datetime
 
@@ -30,7 +29,7 @@ import logging
 log = logging.getLogger(__name__)
 
 
-async def export_project(project, temporary_dir, include_images=False, keep_compute_id=False, allow_all_nodes=False, reset_mac_addresses=False):
+async def export_project(zstream, project, temporary_dir, include_images=False, keep_compute_id=False, allow_all_nodes=False, reset_mac_addresses=False):
     """
     Export a project to a zip file.
 
@@ -53,8 +52,6 @@ async def export_project(project, temporary_dir, include_images=False, keep_comp
     # Make sure we save the project
     project.dump()
 
-    zstream = zipstream.ZipFile(allowZip64=True)
-
     if not os.path.exists(project._path):
         raise aiohttp.web.HTTPNotFound(text="Project could not be found at '{}'".format(project._path))
 
@@ -80,7 +77,7 @@ async def export_project(project, temporary_dir, include_images=False, keep_comp
             if file.endswith(".gns3"):
                 continue
             _patch_mtime(path)
-            zstream.write(path, os.path.relpath(path, project._path), compress_type=zipfile.ZIP_DEFLATED)
+            zstream.write(path, os.path.relpath(path, project._path))
 
     # Export files from remote computes
     downloaded_files = set()
@@ -103,11 +100,9 @@ async def export_project(project, temporary_dir, include_images=False, keep_comp
                     response.close()
                     f.close()
                     _patch_mtime(temp_path)
-                    zstream.write(temp_path, arcname=compute_file["path"], compress_type=zipfile.ZIP_DEFLATED)
+                    zstream.write(temp_path, arcname=compute_file["path"])
                     downloaded_files.add(compute_file['path'])
 
-    return zstream
-
 
 def _patch_mtime(path):
     """
@@ -232,6 +227,7 @@ async def _patch_project_file(project, path, zstream, include_images, keep_compu
     zstream.writestr("project.gns3", json.dumps(topology).encode())
     return images
 
+
 def _export_local_image(image, zstream):
     """
     Exports a local image to the zip file.
diff --git a/gns3server/controller/project.py b/gns3server/controller/project.py
index a7407a69..5ca2d30b 100644
--- a/gns3server/controller/project.py
+++ b/gns3server/controller/project.py
@@ -24,6 +24,7 @@ import shutil
 import asyncio
 import aiohttp
 import tempfile
+import zipfile
 
 from uuid import UUID, uuid4
 
@@ -38,6 +39,7 @@ from ..utils.path import check_path_allowed, get_default_project_directory
 from ..utils.asyncio.pool import Pool
 from ..utils.asyncio import locking
 from ..utils.asyncio import wait_run_in_executor
+from ..utils.asyncio import aiozipstream
 from .export_project import export_project
 from .import_project import import_project
 
@@ -976,9 +978,10 @@ class Project:
         assert self._status != "closed"
         try:
             with tempfile.TemporaryDirectory() as tmpdir:
-                zipstream = await export_project(self, tmpdir, keep_compute_id=True, allow_all_nodes=True, reset_mac_addresses=True)
-                project_path = os.path.join(tmpdir, "project.gns3p")
-                await wait_run_in_executor(self._create_duplicate_project_file, project_path, zipstream)
+                with aiozipstream.ZipFile(compression=zipfile.ZIP_STORED) as zstream:
+                    zipstream = await export_project(zstream, self, tmpdir, keep_compute_id=True, allow_all_nodes=True, reset_mac_addresses=True)
+                    project_path = os.path.join(tmpdir, "project.gns3p")
+                    await wait_run_in_executor(self._create_duplicate_project_file, project_path, zipstream)
                 with open(project_path, "rb") as f:
                     project = await import_project(self._controller, str(uuid.uuid4()), f, location=location, name=name, keep_compute_id=True)
         except (ValueError, OSError, UnicodeEncodeError) as e:
diff --git a/gns3server/handlers/api/controller/project_handler.py b/gns3server/handlers/api/controller/project_handler.py
index 8a1ad4d9..a59802d5 100644
--- a/gns3server/handlers/api/controller/project_handler.py
+++ b/gns3server/handlers/api/controller/project_handler.py
@@ -20,11 +20,14 @@ import sys
 import aiohttp
 import asyncio
 import tempfile
+import zipfile
+import time
 
 from gns3server.web.route import Route
 from gns3server.controller import Controller
 from gns3server.controller.import_project import import_project
 from gns3server.controller.export_project import export_project
+from gns3server.utils.asyncio import aiozipstream
 from gns3server.config import Config
 
 
@@ -301,19 +304,24 @@ class ProjectHandler:
         controller = Controller.instance()
         project = await controller.get_loaded_project(request.match_info["project_id"])
 
+
         try:
+            begin = time.time()
             with tempfile.TemporaryDirectory() as tmp_dir:
-                stream = await export_project(project, tmp_dir, include_images=bool(int(request.query.get("include_images", "0"))))
-                # We need to do that now because export could failed and raise an HTTP error
-                # that why response start need to be the later possible
-                response.content_type = 'application/gns3project'
-                response.headers['CONTENT-DISPOSITION'] = 'attachment; filename="{}.gns3project"'.format(project.name)
-                response.enable_chunked_encoding()
-                await response.prepare(request)
+                with aiozipstream.ZipFile(compression=zipfile.ZIP_DEFLATED) as zstream:
+                    await export_project(zstream, project, tmp_dir, include_images=bool(int(request.query.get("include_images", "0"))))
 
-                for data in stream:
-                    await response.write(data)
+                    # We need to do that now because export could failed and raise an HTTP error
+                    # that why response start need to be the later possible
+                    response.content_type = 'application/octet-stream'
+                    response.headers['CONTENT-DISPOSITION'] = 'attachment; filename="{}.gns3project"'.format(project.name)
+                    response.enable_chunked_encoding()
+                    await response.prepare(request)
 
+                    async for chunk in zstream:
+                        await response.write(chunk)
+
+            log.info("Project '{}' exported in {:.4f} seconds".format(project.id, time.time() - begin))
                 #await response.write_eof() #FIXME: shound't be needed anymore
         # Will be raise if you have no space left or permission issue on your temporary directory
         # RuntimeError: something was wrong during the zip process
diff --git a/gns3server/utils/asyncio/aiozipstream.py b/gns3server/utils/asyncio/aiozipstream.py
new file mode 100644
index 00000000..dfcd49ba
--- /dev/null
+++ b/gns3server/utils/asyncio/aiozipstream.py
@@ -0,0 +1,430 @@
+#!/usr/bin/env python
+#
+# Copyright (C) 2019 GNS3 Technologies Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+"""
+Iterable ZIP archive generator.
+
+Derived directly from zipfile.py and the zipstream project
+https://github.com/allanlei/python-zipstream
+"""
+
+import os
+import sys
+import stat
+import struct
+import time
+import zipfile
+import asyncio
+import aiofiles
+from concurrent import futures
+from async_generator import async_generator, yield_
+
+from zipfile import (structCentralDir, structEndArchive64, structEndArchive, structEndArchive64Locator,
+                     stringCentralDir, stringEndArchive64, stringEndArchive, stringEndArchive64Locator)
+
+stringDataDescriptor = b'PK\x07\x08'  # magic number for data descriptor
+
+
+def _get_compressor(compress_type):
+    """
+    Return the compressor.
+    """
+
+    if compress_type == zipfile.ZIP_DEFLATED:
+        from zipfile import zlib
+        return zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION, zlib.DEFLATED, -15)
+    elif compress_type == zipfile.ZIP_BZIP2:
+        from zipfile import bz2
+        return bz2.BZ2Compressor()
+    elif compress_type == zipfile.ZIP_LZMA:
+        from zipfile import LZMACompressor
+        return LZMACompressor()
+    else:
+        return None
+
+
+class PointerIO(object):
+
+    def __init__(self, mode='wb'):
+        if mode not in ('wb', ):
+            raise RuntimeError('zipstream.ZipFile() requires mode "wb"')
+        self.data_pointer = 0
+        self.__mode = mode
+        self.__closed = False
+
+    @property
+    def mode(self):
+        return self.__mode
+
+    @property
+    def closed(self):
+        return self.__closed
+
+    def close(self):
+        self.__closed = True
+
+    def flush(self):
+        pass
+
+    def next(self):
+        raise NotImplementedError()
+
+    def tell(self):
+        return self.data_pointer
+
+    def truncate(size=None):
+        raise NotImplementedError()
+
+    def write(self, data):
+        if self.closed:
+            raise ValueError('I/O operation on closed file')
+
+        if isinstance(data, str):
+            data = data.encode('utf-8')
+        if not isinstance(data, bytes):
+            raise TypeError('expected bytes')
+        self.data_pointer += len(data)
+        return data
+
+
+class ZipInfo(zipfile.ZipInfo):
+
+    def __init__(self, *args, **kwargs):
+        zipfile.ZipInfo.__init__(self, *args, **kwargs)
+
+    def DataDescriptor(self):
+        """
+        crc-32                          4 bytes
+        compressed size                 4 bytes
+        uncompressed size               4 bytes
+        """
+
+        if self.compress_size > zipfile.ZIP64_LIMIT or self.file_size > zipfile.ZIP64_LIMIT:
+            fmt = b'<4sLQQ'
+        else:
+            fmt = b'<4sLLL'
+        return struct.pack(fmt, stringDataDescriptor, self.CRC, self.compress_size, self.file_size)
+
+
+class ZipFile(zipfile.ZipFile):
+
+    def __init__(self, fileobj=None, mode='w', compression=zipfile.ZIP_STORED, allowZip64=True, chunksize=32768):
+        """Open the ZIP file with mode write "w"."""
+
+        if mode not in ('w', ):
+            raise RuntimeError('aiozipstream.ZipFile() requires mode "w"')
+        if fileobj is None:
+            fileobj = PointerIO()
+
+        self._comment = b''
+        zipfile.ZipFile.__init__(self, fileobj, mode=mode, compression=compression, allowZip64=allowZip64)
+        self._chunksize = chunksize
+        self.paths_to_write = []
+
+    def __aiter__(self):
+        return self._stream()
+
+    @property
+    def comment(self):
+        """
+        The comment text associated with the ZIP file.
+        """
+
+        return self._comment
+
+    @comment.setter
+    def comment(self, comment):
+        """
+        Add a comment text associated with the ZIP file.
+        """
+
+        if not isinstance(comment, bytes):
+            raise TypeError("comment: expected bytes, got %s" % type(comment))
+        # check for valid comment length
+        if len(comment) >= zipfile.ZIP_MAX_COMMENT:
+            if self.debug:
+                print('Archive comment is too long; truncating to %d bytes' % zipfile.ZIP_MAX_COMMENT)
+            comment = comment[:zipfile.ZIP_MAX_COMMENT]
+        self._comment = comment
+        self._didModify = True
+
+    @async_generator
+    async def data_generator(self, path):
+
+        async with aiofiles.open(path, "rb") as f:
+            while True:
+                part = await f.read(self._chunksize)
+                if not part:
+                    break
+                await yield_(part)
+        return
+
+    async def _run_in_executor(self, task, *args, **kwargs):
+        """
+        Run synchronous task in separate thread and await for result.
+        """
+
+        loop = asyncio.get_event_loop()
+        return await loop.run_in_executor(futures.ThreadPoolExecutor(max_workers=1), task, *args, **kwargs)
+
+    @async_generator
+    async def _stream(self):
+
+        for kwargs in self.paths_to_write:
+            async for chunk in self._write(**kwargs):
+                await yield_(chunk)
+        for chunk in self._close():
+            await yield_(chunk)
+
+    def write(self, filename, arcname=None, compress_type=None):
+        """
+        Write a file to the archive under the name `arcname`.
+        """
+
+        kwargs = {'filename': filename, 'arcname': arcname, 'compress_type': compress_type}
+        self.paths_to_write.append(kwargs)
+
+    def write_iter(self, arcname, iterable, compress_type=None):
+        """
+        Write the bytes iterable `iterable` to the archive under the name `arcname`.
+        """
+
+        kwargs = {'arcname': arcname, 'iterable': iterable, 'compress_type': compress_type}
+        self.paths_to_write.append(kwargs)
+
+    def writestr(self, arcname, data, compress_type=None):
+        """
+        Writes a str into ZipFile by wrapping data as a generator
+        """
+
+        def _iterable():
+            yield data
+        return self.write_iter(arcname, _iterable(), compress_type=compress_type)
+
+    @async_generator
+    async def _write(self, filename=None, iterable=None, arcname=None, compress_type=None):
+        """
+        Put the bytes from filename into the archive under the name `arcname`.
+        """
+
+        if not self.fp:
+            raise RuntimeError(
+                  "Attempt to write to ZIP archive that was already closed")
+        if (filename is None and iterable is None) or (filename is not None and iterable is not None):
+            raise ValueError("either (exclusively) filename or iterable shall be not None")
+
+        if filename:
+            st = os.stat(filename)
+            isdir = stat.S_ISDIR(st.st_mode)
+            mtime = time.localtime(st.st_mtime)
+            date_time = mtime[0:6]
+        else:
+            st, isdir, date_time = None, False, time.localtime()[0:6]
+        # Create ZipInfo instance to store file information
+        if arcname is None:
+            arcname = filename
+        arcname = os.path.normpath(os.path.splitdrive(arcname)[1])
+        while arcname[0] in (os.sep, os.altsep):
+            arcname = arcname[1:]
+        if isdir:
+            arcname += '/'
+        zinfo = ZipInfo(arcname, date_time)
+        if st:
+            zinfo.external_attr = (st[0] & 0xFFFF) << 16      # Unix attributes
+        else:
+            zinfo.external_attr = 0o600 << 16     # ?rw-------
+        if compress_type is None:
+            zinfo.compress_type = self.compression
+        else:
+            zinfo.compress_type = compress_type
+
+        if st:
+            zinfo.file_size = st[6]
+        else:
+            zinfo.file_size = 0
+        zinfo.flag_bits = 0x00
+        zinfo.flag_bits |= 0x08                 # ZIP flag bits, bit 3 indicates presence of data descriptor
+        zinfo.header_offset = self.fp.tell()    # Start of header bytes
+        if zinfo.compress_type == zipfile.ZIP_LZMA:
+            # Compressed data includes an end-of-stream (EOS) marker
+            zinfo.flag_bits |= 0x02
+
+        self._writecheck(zinfo)
+        self._didModify = True
+
+        if isdir:
+            zinfo.file_size = 0
+            zinfo.compress_size = 0
+            zinfo.CRC = 0
+            self.filelist.append(zinfo)
+            self.NameToInfo[zinfo.filename] = zinfo
+            await yield_(self.fp.write(zinfo.FileHeader(False)))
+            return
+
+        cmpr = _get_compressor(zinfo.compress_type)
+
+        # Must overwrite CRC and sizes with correct data later
+        zinfo.CRC = CRC = 0
+        zinfo.compress_size = compress_size = 0
+        # Compressed size can be larger than uncompressed size
+        zip64 = self._allowZip64 and zinfo.file_size * 1.05 > zipfile.ZIP64_LIMIT
+        await yield_(self.fp.write(zinfo.FileHeader(zip64)))
+
+        file_size = 0
+        if filename:
+            async for buf in self.data_generator(filename):
+                file_size = file_size + len(buf)
+                CRC = zipfile.crc32(buf, CRC) & 0xffffffff
+                if cmpr:
+                    buf = await self._run_in_executor(cmpr.compress, buf)
+                    compress_size = compress_size + len(buf)
+                await yield_(self.fp.write(buf))
+        else: # we have an iterable
+            for buf in iterable:
+                file_size = file_size + len(buf)
+                CRC = zipfile.crc32(buf, CRC) & 0xffffffff
+                if cmpr:
+                    buf = await self._run_in_executor(cmpr.compress, buf)
+                    compress_size = compress_size + len(buf)
+                    await yield_(self.fp.write(buf))
+
+        if cmpr:
+            buf = cmpr.flush()
+            compress_size = compress_size + len(buf)
+            await yield_(self.fp.write(buf))
+            zinfo.compress_size = compress_size
+        else:
+            zinfo.compress_size = file_size
+        zinfo.CRC = CRC
+        zinfo.file_size = file_size
+        if not zip64 and self._allowZip64:
+            if file_size > zipfile.ZIP64_LIMIT:
+                raise RuntimeError('File size has increased during compressing')
+            if compress_size > zipfile.ZIP64_LIMIT:
+                raise RuntimeError('Compressed size larger than uncompressed size')
+
+        await yield_(self.fp.write(zinfo.DataDescriptor()))
+        self.filelist.append(zinfo)
+        self.NameToInfo[zinfo.filename] = zinfo
+
+    def _close(self):
+        """
+        Close the file, and for mode "w" write the ending records.
+        """
+
+        if self.fp is None:
+            return
+
+        try:
+            if self.mode in ('w', 'a') and self._didModify:  # write ending records
+                count = 0
+                pos1 = self.fp.tell()
+                for zinfo in self.filelist:         # write central directory
+                    count = count + 1
+                    dt = zinfo.date_time
+                    dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2]
+                    dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2)
+                    extra = []
+                    if zinfo.file_size > zipfile.ZIP64_LIMIT or zinfo.compress_size > zipfile.ZIP64_LIMIT:
+                        extra.append(zinfo.file_size)
+                        extra.append(zinfo.compress_size)
+                        file_size = 0xffffffff
+                        compress_size = 0xffffffff
+                    else:
+                        file_size = zinfo.file_size
+                        compress_size = zinfo.compress_size
+
+                    if zinfo.header_offset > zipfile.ZIP64_LIMIT:
+                        extra.append(zinfo.header_offset)
+                        header_offset = 0xffffffff
+                    else:
+                        header_offset = zinfo.header_offset
+
+                    extra_data = zinfo.extra
+                    min_version = 0
+                    if extra:
+                        # Append a ZIP64 field to the extra's
+                        extra_data = struct.pack(
+                                b'<HH' + b'Q'*len(extra),
+                                1, 8*len(extra), *extra) + extra_data
+                        min_version = zipfile.ZIP64_VERSION
+
+                    if zinfo.compress_type == zipfile.ZIP_BZIP2:
+                        min_version = max(zipfile.BZIP2_VERSION, min_version)
+                    elif zinfo.compress_type == zipfile.ZIP_LZMA:
+                        min_version = max(zipfile.LZMA_VERSION, min_version)
+
+                    extract_version = max(min_version, zinfo.extract_version)
+                    create_version = max(min_version, zinfo.create_version)
+                    try:
+                        filename, flag_bits = zinfo._encodeFilenameFlags()
+                        centdir = struct.pack(structCentralDir,
+                            stringCentralDir, create_version,
+                            zinfo.create_system, extract_version, zinfo.reserved,
+                            flag_bits, zinfo.compress_type, dostime, dosdate,
+                            zinfo.CRC, compress_size, file_size,
+                            len(filename), len(extra_data), len(zinfo.comment),
+                            0, zinfo.internal_attr, zinfo.external_attr,
+                            header_offset)
+                    except DeprecationWarning:
+                        print((structCentralDir, stringCentralDir, create_version,
+                            zinfo.create_system, extract_version, zinfo.reserved,
+                            zinfo.flag_bits, zinfo.compress_type, dostime, dosdate,
+                            zinfo.CRC, compress_size, file_size,
+                            len(zinfo.filename), len(extra_data), len(zinfo.comment),
+                            0, zinfo.internal_attr, zinfo.external_attr,
+                            header_offset), file=sys.stderr)
+                        raise
+                    yield self.fp.write(centdir)
+                    yield self.fp.write(filename)
+                    yield self.fp.write(extra_data)
+                    yield self.fp.write(zinfo.comment)
+
+                pos2 = self.fp.tell()
+                # Write end-of-zip-archive record
+                centDirCount = count
+                centDirSize = pos2 - pos1
+                centDirOffset = pos1
+                if (centDirCount >= zipfile.ZIP_FILECOUNT_LIMIT or
+                    centDirOffset > zipfile.ZIP64_LIMIT or
+                    centDirSize > zipfile.ZIP64_LIMIT):
+                    # Need to write the ZIP64 end-of-archive records
+                    zip64endrec = struct.pack(
+                            structEndArchive64, stringEndArchive64,
+                            44, 45, 45, 0, 0, centDirCount, centDirCount,
+                            centDirSize, centDirOffset)
+                    yield self.fp.write(zip64endrec)
+
+                    zip64locrec = struct.pack(
+                            structEndArchive64Locator,
+                            stringEndArchive64Locator, 0, pos2, 1)
+                    yield self.fp.write(zip64locrec)
+                    centDirCount = min(centDirCount, 0xFFFF)
+                    centDirSize = min(centDirSize, 0xFFFFFFFF)
+                    centDirOffset = min(centDirOffset, 0xFFFFFFFF)
+
+                endrec = struct.pack(structEndArchive, stringEndArchive,
+                                    0, 0, centDirCount, centDirCount,
+                                    centDirSize, centDirOffset, len(self._comment))
+                yield self.fp.write(endrec)
+                yield self.fp.write(self._comment)
+                self.fp.flush()
+        finally:
+            fp = self.fp
+            self.fp = None
+            if not self._filePassed:
+                fp.close()
diff --git a/requirements.txt b/requirements.txt
index 9b6b5904..36bb8b4a 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,10 +1,11 @@
 jsonschema>=2.4.0
 aiohttp==3.5.4
 aiohttp-cors==0.7.0
+aiofiles==0.4.0
+async_generator>=1.10
 Jinja2>=2.7.3
 raven>=5.23.0
 psutil>=3.0.0
-zipstream>=1.1.4
 prompt-toolkit==1.0.15
 async-timeout==3.0.1
 distro>=1.3.0