0

We're releasing a Messenger Platform ChatBot on a page with approx. 130k fans soon. At the moment we host on an EC2 instance with 2 Cores and 4GB RAM, which runs smoothly with our 30 users atm.

Our tech stack is:

NGINX gunicorn (4 workers) Flask-framework DB is separated on amazon RDS (Postgres)

Do you think our server can handle the first wave of requests after launch? Can't imagine how big this is going to be.

Any advice regarding hardware? Should be switch to a bigger instance or maybe Use the AWS loadbalancer to spin up a second instance on heavy load?

Thanks in advance!

trnc
  • 688
  • 1
  • 12
  • 30
  • No way to know. Depends enormously on how efficient your code is. Use something like https://www.blitz.io/ to simulate more realistic amounts of load. – ceejayoz May 18 '17 at 20:25
  • 1
    You should use an ELB and auto scaling based on whatever resource you're likely to run out of, probably CPU. Once you see it's working you don't have to worry about it, it will scale up and down as required. If your server is often well below capacity you may even be able to reduce your instance size, it'll just spin up as many instances as are required. Be careful with t2 instances behind a load balancer. – Tim May 18 '17 at 21:00

0 Answers0