perf(python): prealloc size for dict

apache · Nov 20, 2024 · 06e5d84 · 06e5d84
1 parent 8add13c
commit 06e5d84
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 3 deletions.
diff --git a/integration_tests/cpython_benchmark/fury_benchmark.py b/integration_tests/cpython_benchmark/fury_benchmark.py
@@ -64,6 +64,7 @@
     "view_count": 7,
     "zip": "",
 }
+LARGE_DICT = {str(i) : i for i in range(2**10 + 1)}
 
 TUPLE = (
     [
@@ -177,6 +178,7 @@ def micro_benchmark():
     runner.parse_args()
     language = pyfury.Language.XLANG if args.xlang else pyfury.Language.PYTHON
     runner.bench_func("fury_dict", fury_object, language, not args.no_ref, DICT)
+    runner.bench_func("fury_large_dict", fury_object, language, not args.no_ref, LARGE_DICT)
     runner.bench_func(
         "fury_dict_group", fury_object, language, not args.no_ref, DICT_GROUP
     )

diff --git a/python/pyfury/_serialization.pyx b/python/pyfury/_serialization.pyx
@@ -76,6 +76,7 @@ cdef extern from *:
     """
     object int2obj(int64_t obj_addr)
     int64_t obj2int(object obj_ref)
+    dict _PyDict_NewPresized(Py_ssize_t minused)
 
 
 cdef int8_t NULL_FLAG = -3
@@ -2081,9 +2082,9 @@ cdef class MapSerializer(Serializer):
     cpdef inline read(self, Buffer buffer):
         cdef MapRefResolver ref_resolver = self.ref_resolver
         cdef ClassResolver class_resolver = self.class_resolver
-        cdef dict map_ = {}
-        ref_resolver.reference(map_)
         cdef int32_t len_ = buffer.read_varint32()
+        cdef dict map_ = _PyDict_NewPresized(len_)
+        ref_resolver.reference(map_)
         cdef int32_t ref_id
         cdef ClassInfo key_classinfo
         cdef ClassInfo value_classinfo
@@ -2131,7 +2132,7 @@ cdef class MapSerializer(Serializer):
 
     cpdef inline xread(self, Buffer buffer):
         cdef int32_t len_ = buffer.read_varint32()
-        cdef dict map_ = {}
+        cdef dict map_ = _PyDict_NewPresized(len_)
         self.fury.ref_resolver.reference(map_)
         for i in range(len_):
             k = self.fury.xdeserialize_ref(