DB used : Elasticsearch as a value can be searched and indexed on the basis of mathching as well(we might have to find a word in a paragraph, or a word closest to a particular word).
- Couldn't use SQL databases as the best case would be to traverse throught all entry and search for a word in a column.
- Didn't use MongoDB as search a value in multi_match is easy in elasticsearch and we get a ranking of data closest to the query.
Language Used : Golang (highly scalable language,easy to learn and a fast and easy to learn language.
Error Tracing, Logs, track performance, Error alerts : Sentry because of easy GoLang integration, complete context, real time updates.
Queue Service : Used Kafka Services, because of producer and consumer configuration, partition and task that make it scalable.
- Didn't use RabbitMQ as it doesn't support multiple consumers.
