Skip to main content

Posts

Solution for DB Query on Multi DB Servers

In HMS why do we need this tool? To increase the security as well as efficiency, even though using MySQL as the main DB, we divide them into 6 separate MySQL Servers Each Server contains one or more Databases There is always a need to join cross dbs between different servers Right now, what we did are: a. Use App APIs to query and then store in a temp table in the same server and then join --> Microserver Best Practice but it takes a lot of devs and codes b. Extract/Pull all of them to a datalake ( GCP Bigquery for now) and then query from there --> Correct solution for a Big Data Concept but its not match for a real time request and it costly if we want to do that. So what we need is a way to query from multiple database source and joinable . Morever, if its SQL-like, it will offload a lot of dev tasks and share them with Data Analysts By query directly on the Source without ETL, we can guarrantee the REALTIME as well as maximize the cost ( coz we can do them ON-DEMAND...
Recent posts

Solution for DB Query on Multi DB Servers

In HMS why do we need this tool? To increase the security as well as efficiency, even though using MySQL as the main DB, we divide them into 6 separate MySQL Servers Each Server contains one or more Databases There is always a need to join cross dbs between different servers Right now, what we did are: a. Use App APIs to query and then store in a temp table in the same server and then join --> Microserver Best Practice but it takes a lot of devs and codes b. Extract/Pull all of them to a datalake ( GCP Bigquery for now) and then query from there --> Correct solution for a Big Data Concept but its not match for a real time request and it costly if we want to do that. So what we need is a way to query from multiple database source and joinable . Morever, if its SQL-like, it will offload a lot of dev tasks and share them with Data Analysts By query directly on the Source without ETL, we can guarrantee the REALTIME as well as maximize the cost ( coz we can do them ON-DEMAND...