Each isolation level in chronological order

Tiger's Place

This post provides a comprehensive explanation of database ACID isolation levels and the associated race conditions. It covers isolation levels from weakest to strongest, detailing their characteristics and issues, including dirty reads, non-repeatable reads, write skew, and phantoms. Solutions such as snapshot isolation, two-phase locking (2PL), and Serializable Snapshot Isolation (SSI) are also discussed along with practical examples.

All database ACID isolation levels and race conditions explained

Permalink Three Ways to achieve serializability

Really good article explained a lot of things in a good structural manner.
We a state-full server and we maintain a lot of data in memory and we face most of the issues mentioned here, and our team slowly built these solutions only but I never thought it like this way, now I’m seeing it like we have created a in-memory db.
Like we are using read and write locks approach because we could have a lot of conflicts since the data has frequent updates. But after reading this article I want to go an evaluate how frequent the data updates are and can we go ahead with SSI method.

This is an extremely usefull article which has opened me the eyes on many things.
Thanks