Lernen sys.getsizeof() and Object Size Inspection

Swipe um das Menü anzuzeigen

sys.getsizeof() is the most direct tool for measuring how much memory a single Python object occupies. Understanding what it measures – and what it does not – prevents common misinterpretations and helps you accurately compare the memory cost of different data representations.

What `sys.getsizeof()` Measures

sys.getsizeof() returns the size of the object itself in bytes – the object header, internal pointers, and any inline data. It does not include the size of objects the target object references:


              1234567891011
            
import sys

# getsizeof measures the container, not its contents
price_list = [1500.0, 2300.0, 4100.0, 750.0]
print(sys.getsizeof(price_list))       # ~120 bytes – list structure + 4 pointers

single_float = 1500.0
print(sys.getsizeof(single_float))     # 24 bytes – one float object

# The list does NOT include the floats it points to
print(sys.getsizeof(price_list) + 4 * sys.getsizeof(1500.0))  # Closer to true total

Total Size Including References

To measure the true memory footprint of a nested structure, recursively sum all referenced objects:


              12345678910111213141516171819202122232425262728
            
import sys

def total_size(obj, seen=None):
    """Recursively calculate total memory of an object and all it references."""
    size = sys.getsizeof(obj)
    if seen is None:
        seen = set()
    obj_id = id(obj)
    if obj_id in seen:
        return 0
    seen.add(obj_id)

    if isinstance(obj, dict):
        size += sum(total_size(key, seen) + total_size(value, seen) for key, value in obj.items())
    elif isinstance(obj, (list, tuple, set, frozenset)):
        size += sum(total_size(item, seen) for item in obj)

    return size

# Comparing shallow vs total size
transaction_record = {
    "id": "TX_001",
    "amount": 4500.0,
    "tags": ["urgent", "international", "flagged"],
}

print(f"getsizeof only:   {sys.getsizeof(transaction_record)} bytes")
print(f"Total recursive:  {total_size(transaction_record)} bytes")

Comparing Common Structures


              1234567891011
            
import sys
import array
from collections import deque

items = list(range(1000))

print(f"list of 1000 ints:      {sys.getsizeof(items)} bytes (container only)")
print(f"tuple of 1000 ints:     {sys.getsizeof(tuple(items))} bytes")
print(f"set of 1000 ints:       {sys.getsizeof(set(items))} bytes")
print(f"array.array of 1000:    {sys.getsizeof(array.array('i', items))} bytes")
print(f"deque of 1000 ints:     {sys.getsizeof(deque(items))} bytes")

array.array stores values inline – sys.getsizeof() reflects the actual data size. For list, tuple, and set, the reported size is the container overhead plus pointers – the integer objects themselves are counted separately.

String Size Breakdown

Strings in Python carry encoding overhead. The size scales with content:


              1234567
            
import sys

print(sys.getsizeof(""))           # 49 bytes – empty string overhead
print(sys.getsizeof("a"))          # 50 bytes – one ASCII character
print(sys.getsizeof("hello"))      # 54 bytes – 5 ASCII characters
print(sys.getsizeof("héllo"))      # Larger – non-ASCII forces UCS-2 or UCS-4 encoding
print(sys.getsizeof("hello" * 100))  # ~549 bytes – 500 characters

Practical Size Reference

War alles klar?

Danke für Ihr Feedback!

Abschnitt 4. Kapitel 4

Fragen Sie AI

Fragen Sie alles oder probieren Sie eine der vorgeschlagenen Fragen, um unser Gespräch zu beginnen