1

So I'm wondering, currently we use Digital Ocean, to host our kubernetes cluster.

They do offer 3 types of loadbalancers: small, medium, large.

You can reference them here. They state that for the large loadbalancer they have "Maximum Simultaneous Connections: 40,000" and "Maximum New SSL Connections per Second: 1000".

So if you e.g. need more than 40,000 simultaneous connections or more SSL connections per second? How can you handle such requirments, would you add multiple Loadbalancers and have multiple A records pointing to them? How do big companies handle this?

Having only 40,000 simultaneous connections is ab bottleneck.

natschz
  • 1,007
  • 10
  • 23
  • possible duplicate of: https://stackoverflow.com/questions/55201050/what-can-we-do-when-load-balancer-becomes-the-bottleneck – Krishna Chaurasia Jan 18 '21 at 13:24
  • Yea thats kind of what I was thinking about, but I am wondering what the implications are especially concerning a kubernetes cluster. How does this play with ingresses and can I just add multiple loadbalancer. – natschz Jan 18 '21 at 13:32
  • I think this is related to specifically to Digital Ocean, not a general question, so it's not a duplicate in my opinion. – BvuRVKyUVlViVIc7 Jan 18 '21 at 13:32

0 Answers0