Use Zarr v2.13.0a2 (#129)

which include zarr-developers/zarr-python#934

Authors:
  - Mads R. B. Kristensen (https://github.com/madsbk)

Approvers:
  - Benjamin Zaitlen (https://github.com/quasiben)

URL: #129

python/benchmarks/single-node-io.py

@@ -3,7 +3,6 @@
 
 import argparse
 import contextlib
-import functools
 import os
 import os.path
 import pathlib
@@ -193,30 +192,30 @@ def run_posix(args):
     return read_time, write_time
 
 
-def run_zarr(store_type, args):
+def run_zarr(args):
     """Use the Zarr API"""
 
     import zarr
-    import zarr.cupy
 
     import kvikio.zarr
 
+    dir_path = args.dir / "zarr"
+
+    if not hasattr(zarr.Array, "meta_array"):
+        raise RuntimeError("requires Zarr v2.13+")
+
     a = cupy.arange(args.nbytes, dtype="uint8")
 
-    # Retrieve the store and compressor to use based on `store_type`
-    shutil.rmtree(str(args.dir / store_type), ignore_errors=True)
-    store, compressor = {
-        "gds": (kvikio.zarr.GDSStore(args.dir / store_type), None),
-        "posix": (
-            zarr.DirectoryStore(args.dir / store_type),
-            zarr.cupy.CuPyCPUCompressor(),
-        ),
-    }[store_type]
+    shutil.rmtree(str(dir_path), ignore_errors=True)
 
     # Write
     t0 = clock()
     z = zarr.array(
-        a, chunks=False, compressor=compressor, store=store, meta_array=cupy.empty(())
+        a,
+        chunks=False,
+        compressor=None,
+        store=kvikio.zarr.GDSStore(dir_path),
+        meta_array=cupy.empty(()),
     )
     write_time = clock() - t0
 
@@ -231,8 +230,7 @@ def run_zarr(store_type, args):
 
 API = {
     "cufile": run_cufile,
-    "zarr-gds": functools.partial(run_zarr, "gds"),
-    "zarr-posix": functools.partial(run_zarr, "posix"),
+    "zarr": run_zarr,
     "posix": run_posix,
     "cufile-mfma": run_cufile_multiple_files_multiple_arrays,
     "cufile-mf": run_cufile_multiple_files,

python/kvikio/zarr.py

@@ -1,15 +1,11 @@
 # Copyright (c) 2021-2022, NVIDIA CORPORATION. All rights reserved.
 # See file LICENSE for terms.
 
-import errno
 import os
 import os.path
-import shutil
-import uuid
 
 import cupy
 import zarr.storage
-from zarr.util import retry_call
 
 import kvikio
 from kvikio._lib.arr import asarray
@@ -55,59 +51,3 @@ def _tofile(self, a, fn):
                 assert written == a.nbytes
         else:
             super()._tofile(a.obj, fn)
-
-    def __setitem__(self, key, value):
-        """
-        We have to overwrite this because `DirectoryStore.__setitem__`
-        converts `value` to a NumPy array always
-        """
-        key = self._normalize_key(key)
-
-        # coerce to flat, contiguous buffer (ideally without copying)
-        arr = asarray(value)
-        if arr.contiguous:
-            value = arr
-        else:
-            if arr.cuda:
-                # value = cupy.ascontiguousarray(value)
-                value = arr.reshape(-1, order="A")
-            else:
-                # can flatten without copy
-                value = arr.reshape(-1, order="A")
-
-        # destination path for key
-        file_path = os.path.join(self.path, key)
-
-        # ensure there is no directory in the way
-        if os.path.isdir(file_path):
-            shutil.rmtree(file_path)
-
-        # ensure containing directory exists
-        dir_path, file_name = os.path.split(file_path)
-        if os.path.isfile(dir_path):
-            raise KeyError(key)
-        if not os.path.exists(dir_path):
-            try:
-                os.makedirs(dir_path)
-            except OSError as e:
-                if e.errno != errno.EEXIST:
-                    raise KeyError(key)
-
-        # write to temporary file
-        # note we're not using tempfile.NamedTemporaryFile to avoid
-        # restrictive file permissions
-        temp_name = file_name + "." + uuid.uuid4().hex + ".partial"
-        temp_path = os.path.join(dir_path, temp_name)
-        try:
-            self._tofile(value, temp_path)
-
-            # move temporary file into place;
-            # make several attempts at writing the temporary file to get past
-            # potential antivirus file locking issues
-            retry_call(
-                os.replace, (temp_path, file_path), exceptions=(PermissionError,)
-            )
-        finally:
-            # clean up if temp file still exists for whatever reason
-            if os.path.exists(temp_path):  # pragma: no cover
-                os.remove(temp_path)

python/tests/test_benchmarks.py

@@ -21,21 +21,16 @@
         "cufile-mfma",
         "cufile-mf",
         "cufile-ma",
-        "zarr-gds",
-        "zarr-posix",
+        "zarr",
     ],
 )
 def test_single_node_io(run_cmd, tmp_path, api):
     """Test benchmarks/single-node-io.py"""
 
     if "zarr" in api:
-        pytest.importorskip(
-            "zarr.cupy",
-            reason=(
-                "To use Zarr arrays with GDS directly, Zarr needs CuPy support: "
-                "<https://github.com/zarr-developers/zarr-python/pull/934>"
-            ),
-        )
+        zarr = pytest.importorskip("zarr")
+        if not hasattr(zarr.Array, "meta_array"):
+            pytest.skip("requires Zarr v2.13+")
 
     retcode = run_cmd(
         cmd=[

python/tests/test_zarr.py

@@ -8,6 +8,11 @@
 zarr = pytest.importorskip("zarr")
 GDSStore = pytest.importorskip("kvikio.zarr").GDSStore
 
+# To support CuPy arrays, we need the `meta_array` argument introduced in
+# Zarr v2.13, see <https://github.com/zarr-developers/zarr-python/pull/934>
+if not hasattr(zarr.Array, "meta_array"):
+    pytest.skip("requires Zarr v2.13+", allow_module_level=True)
+
 
 @pytest.fixture
 def store(tmp_path):
@@ -32,14 +37,6 @@ def test_direct_store_access(store, array_type):
 def test_array(store):
     """Test Zarr array"""
 
-    pytest.importorskip(
-        "zarr.cupy",
-        reason=(
-            "To use Zarr arrays with GDS directly, Zarr needs CuPy support: "
-            "<https://github.com/zarr-developers/zarr-python/pull/934>"
-        ),
-    )
-
     a = cupy.arange(100)
     z = zarr.array(
         a, chunks=10, compressor=None, store=store, meta_array=cupy.empty(())
@@ -53,14 +50,6 @@ def test_array(store):
 def test_group(store):
     """Test Zarr group"""
 
-    pytest.importorskip(
-        "zarr.cupy",
-        reason=(
-            "To use Zarr arrays with GDS directly, Zarr needs CuPy support: "
-            "<https://github.com/zarr-developers/zarr-python/pull/934>"
-        ),
-    )
-
     g = zarr.open_group(store, meta_array=cupy.empty(()))
     g.ones("data", shape=(10, 11), dtype=int, compressor=None)
     a = g["data"]