Keeping object in memory between api call's in Python Falcon

Question

I'm developing a backend api service using python Falcon. In order to serve each api call, I need to use an object (same object for all requests) which is being refreshed every X hours (let's say 1 hour). Currently this object is saved as pickle in S3. The problem is, that the pickle is quite big (~20 MB) so reading the pickle for every api call seem not efficient. However I'm not sure what is the right approach to keep this object in- memory between different api calls, and how to refresh it every X hours. Because the pickle is relatively big, I don't want to store it locally, and prefer an in-memory shared object approach.

Thx, Oren

If it's that big, have you considered just using a database? — Artyer, Aug 10 '17 at 13:29
It's not very big, but as I said I prefer an in-memory solution as part of my own python process (not additional in-memory db) — Oren Razon, Aug 10 '17 at 13:50
Yes, I attach a link that explains how to persist a DB connection, which is basically what we are trying to do here https://falcon.readthedocs.io/en/stable/user/faq.html#how-do-i-manage-my-database-connections — bluesummers, Apr 23 '18 at 08:43

score 0 · Accepted Answer · answered Aug 11 '17 at 02:37

0

I am not sure weither I understand your requirements exactly or not. But as you want to store in memory only and part of same python process, you can think about global variables and use it. Refer this Using global variables in a function other than the one that created them

It may help you.

answered Aug 11 '17 at 02:37

Harshad Kavathiya

8,939
1
10
19

Hi, actually it solved my problem exactly. Thank u very much! – Oren Razon Jul 24 '18 at 07:03

Keeping object in memory between api call's in Python Falcon

1 Answers1