Does anybody know, how resilience could be achieved if any of your microservices is down. Please suggest me if any. I had one approach in mind but that doesn't seem to be efficient.

Doesn't an answer to this depend on your architecture as a whole?

Yea, that's a pretty broadly scoped question. What are you using for messaging queues, retries and error handling?

First thing I thought about is that, you could have message broker's or some sort of load balance between them, and also, each service would have replicates of it in different machines, this way, whenever one goes down there will be another one to serve to the same purpose. Search a bit about PaaS, automatic scaling, open shift, docker, kubernetes and twelve factor apps