Is it an atipattern to put a function with closure to ray object storage

1

In ray if we have a function as below this is an antipattern as we dont put the big object to object storage.

bigobject=df

@ray.remote
def remote1():
   return len(bigobject)

But if we create a normal python function which does the same logic. And put this normal function to object storage, wouldnt the bigobject in closure also go to object storage. So from a different remote function can we just execute our normal function without causing any performance issues? What would be the drawbacks of this approach in comarison to the suggestions given in Anti-pattern: Closure capturing large objects harms performance — Ray 2.9.0

@ray.remote
def remote2(func)
	return func()

bigobject=df
def func():
	return len(bigobject)

funcObj=ray.put(func)

remote2.remote(funcObj)