In any database system, whenever a client is trying to update a record, it acquires a lock in that record so that another client cannot update the same record. This is done to maintain consistency in RDBMS.

Sometimes, when we perform a bulk update, this locking of records escalates to locking a table, and all access is denied to that particular table. This has an adverse impact on the application's performance. A majority of the application performance issues that arise in the production environment happen due to locking of records or tables. There are various levels of locking in RDBMS depending on the application logic requirement. In my experience as a SQL DBA, I have often observed queries getting timed out while waiting to get locked on a particular record, which is already being locked by another query impacting the application's performance. I have spent a lot of time determining locks that a particular query has on a table to understand its impact on performance...