How python saves memory for creating a large number of instances is as follows
Case in point:
In an online game, the player class Player(id, name, status,....) is defined. For every online player, there is an instance of Player within the server program, and when there are a lot of online players, a lot of instances (millions) will be created.
Demand:
How can we reduce the memory overhead of these large numbers of instances?
How do I do it?
First of all, it is important to understand that classes in python can dynamically add attributes, in fact, there is a __dict__ method in memory that maintains such dynamically added attributes, and it occupies memory, so if you turn it off, won't you be able to achieve the memory-saving requirement?
#!/usr/bin/python3
import time
import sys
class Player(object):
def __init__(self, id, name, status):
= id
= name
= status
if __name__ == '__main__':
player_1 = Player(1, 'bei_bei', 'Going live')
print(player_1.__dict__)
print('_' * 100)
# Dynamic assembly properties
player_1.money = 10000
player_1.__dict__['time'] = ()
print(player_1.__dict__)
print(player_1.money, player_1.time)
print('_' * 100)
# Print the amount of memory space occupied by __dict__.
print('The dict method takes up memory:', (player_1.__dict__))
print('_'*100)
# Dynamically remove attributes
print(player_1.__dict__)
del player_1.__dict__['time']
del player_1.money
print(player_1.__dict__)
Declare a list of instance attribute names via the __slots__ attribute
#!/usr/bin/python3
class Player(object):
# Specify the fixed-length attribute of a class via the slots method
__slots__ = ['id', 'name', 'status']
def __init__(self, id, name, status):
= id
= name
= status
if __name__ == '__main__':
player_1 = Player(1, 'bei_bei', 'Going live')
print(player_1.id, player_1.name, player_1.status)
# Try to output the __dict__ attribute and find that without it, you can't dynamically assemble class attributes to save memory
try:
print(player_1.__dict__)
except Exception as e:
print(e)
This is the whole content of this article.