# Why the Author Prefers Relational Databases Over Many NoSQL Systems

DevFeed: [Why the Author Prefers Relational Databases Over Many NoSQL Systems](<https://devfeed.tech/articles/postgres-the-death-of-nosql-41105.md>)

Original publisher: [Read original article](<https://www.craigkerstiens.com/2011/11/08/Postgres...-The-death-of-NoSQL/>)

Author: Map

Published: 2011-11-08T20:55:56Z

Content type: opinion

Language: en

Sources: [Craig Kerstiens](<https://devfeed.tech/sources/craig-kerstiens.md>)

Topics: [NoSQL](<https://devfeed.tech/topics/nosql.md>), [acid](<https://devfeed.tech/topics/acid.md>), [Databases](<https://devfeed.tech/topics/databases.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [Replication](<https://devfeed.tech/topics/replication.md>), [backups](<https://devfeed.tech/topics/backups.md>), [MySQL](<https://devfeed.tech/topics/mysql.md>), [Redis](<https://devfeed.tech/topics/redis.md>)

Tags: [acid](<https://devfeed.tech/tags/acid.md>), [backups](<https://devfeed.tech/tags/backups.md>), [data](<https://devfeed.tech/tags/data.md>), [databases](<https://devfeed.tech/tags/databases.md>), [mysql](<https://devfeed.tech/tags/mysql.md>), [nosql](<https://devfeed.tech/tags/nosql.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [redis](<https://devfeed.tech/tags/redis.md>), [relational-databases](<https://devfeed.tech/tags/relational-databases.md>), [replication](<https://devfeed.tech/tags/replication.md>), [sql](<https://devfeed.tech/tags/sql.md>), [transactions](<https://devfeed.tech/tags/transactions.md>)

## AI overview

The article argues that relational databases are preferable as primary data stores when strong data guarantees matter. It highlights ACID properties, SQL as a broadly shared standard, and the portability of SQL-related skills, while acknowledging that some NoSQL systems can offer speed and may suit exceptions such as reporting applications.

## Source excerpt

NoSQL has long been a trend that many have talked about. While there's a place for various key-value stores and tools such as memcache and redis, this will address most specifically how NoSQL is attempting to replace a traditional database. I've long been a fan of postgres and in general traditional relational databases. In a broad sense traditional databases offer multiple things. RDMS Data guarantees The current major SQL databases (SQL Server, Postgres, MySQL, Oracle) offer guarantees around your data that doesn't always exist with other systems. At a very high level this means when they say they have the data there's not a chance they'll loose it. When using something as a primary datastore this is always my first requirement. Data is a valuable commodity so keeping it around is obviously important. There are cases where exceptions exist (reporting applications are common here). The specific thing I always look for is that a system upholds the ACID properties. For a quick breakdown of these: A is for atomic. In short it means no transaction can be partially completed, its all or nothing. C is for consistent. This means you go from one consistent state to another. Meaning things like cascades and constraints are upheld and can't be ignored for a period of time. I is for isolation. This means transactions don't get to interfere with each other. D is for durability. This means once the transactions there its not going anywhere. These basic principles make me feel pretty content with my data being safe. This doesn't include things like backups and replication, but rather is a baseline for me feeling safe with a system. Here's a hint, many NoSQL solutions don't enforce these which is where they get speed from Consistent means for accessing data (SQL) Many people complain about SQL and while its not a perfect language it is a common standard for accessing data. There are idioms that exist in Oracle that do not in Postgres and ones that exist in MySQL that do not in SQ