Golang

If 1000 concurrent requests come, will there be 1000 OS threads? What will happen to the service in this case?

You have a facade that aggregates data from 10 services, each responds in about 100 ms, totaling about a second. How would you solve this problem to make the data return faster?

If the first of 11 services returns an error, will you continue to access the other services for data?

Was Redis in one data center, and everyone accessed it?

On what principle did you choose the architecture and service boundaries?

How to solve the data race problem we discussed?

Was Redis present in all three data centers?

What are possible side effects when using goroutines for file processing, and what should one be prepared for?

You had three data centers — how was the replication/reservation scheme organized between them?

Were data centers full copies of each other, or did they complement each other? Did you just route traffic between them?

What if goroutines write to the same string/buffer — is such a problem possible?

What kind of lock field do you mean (clarification on denormalization, is_blocked flag)?

How can goroutines 'leak'?

How did the data get to all three data centers when an event arrives in the first data center from Kafka, and we catch it and store it in Redis?

Are there any other ways besides channels to organize work with goroutines?

You wrote in your resume that you optimized SQL queries. How do you think an execution plan for selecting unblocked sellers (EXPLAIN) would look?

Можешь рассказать про алгоритм аудирования (сверки данных)? Ты принимал в нём участие? Как оно происходило?

Your resume indicates 50-80 thousand RPS. What kind of service was this, and what did it do?

You mentioned read-model with Redis. Was Redis replicated between data centers?

If one of the services you use returns an error, what will you do and how will you handle it if there is a channel?