Database expert on why NoSQL mattered – and SQL still matters

https://medium.com/s-c-a-l-e/database-guru-on-why-nosql-matters-and-sql-still-matters-c64239fe84fd

by jasondc • 11 years ago

132 37 11 years ago

applecore• 11 years ago

It's pretty clear that SQL won, although ideas from “NoSQL” have been assimilated, i.e., storing and querying semi-structured or unstructured data in an otherwise relational data model.

threeseed3• 11 years ago

You are half correct.

SQL itself is finding itself in vogue again. Not full ANSI SQL but derivatives/half implemented equivalents e.g. CQL, HiveQL, Spark SQL, Presto. The type that this article refers to (AMPLab are the original creators of Spark).

But SQL databases themselves are definitely less popular. Mainly as (a) data grows beyond their capability, (b) data becomes more streaming/real time and (c) microservices pushes data into smaller disparate databases. The one big vertically scaled database at the hub of everything is definitely disappearing.

angrybits4• 11 years ago

The one big vertically scaled database at the hub of everything is definitely disappearing.

Maybe in Startupville, CA. But I think you're forgetting that it's a big world out there and there are lots of systems that are built on vertically scaled relational engines that practically print money. The vast majority of companies out there do not have Twitter-scale engineering problems to solve.

tracker12• 11 years ago

I wouldn't say that.. as much as they disgust me, Oracle isn't going anywhere.. and neither are MS-SQL, mySQL/MariaDB, PostgreSQL, SQLite, Firebird, DB2, and a bunch of others.

SQL allows for very generalized tools to be used to query data, aside from whatever application front ends get developed. This is simply harder with other database engines. Not to mention that it's a simple good enough tool in many use cases. Not everyone has twitter/facebook/google sized problems. Most applications can do just fine scaling up. And often too much importance is placed in hitting five nines of availability.

I really like NoSQL databases, flexible schemaless systems and horizontal scaling. That said, you can go a long way with SQL, it's just you can't solve some problems on multiple systems, and you can't solve some problems on a single system... depends on your needs.

ahachete3• 11 years ago

Those derivatives are, apart from not being really SQL (but a SQLish-like in the best case), only target an incredibly small subset of what is SQL.

When they say "we implement a subset of SQL" they really mean "we have a language similar to what used to be a small subset of SQL 92". And let me reinforce that 92, from the times of Windows 3.1. It cannot be compared to today's Modern SQL (http://use-the-index-luke.com/blog/2015-02/modern-sql) found on modern relational databases, like PostgreSQL.

jamiesonbecker• 11 years ago

SQL did win. Sorta.

NoSQL brought a lot of new ideas. Redis might seem superficially like NDB (MySQL Cluster), but underneath, it's extremely different. What kind of sane person could design a single-threaded, single process database? @antirez built something radical that delivers unbelievable performance.

More innovation and invention has happened in data in the last decade than in the previous thirty, and it's because people experimented and weren't afraid to fail.

Most NoSQL datastores swap limitations in one area for advances in another. Riak, Redis, DynamoDB.. S3 is probably one of the most successful examples of NoSQL, and its very limitations are crucial to its success. Each has clearly defined use cases.

Some people say "Big Data" or "NoSQL" when they really just mean Hadoop or MapReduce, but NoSQL just literally means "everything that's not SQL."

SQL is still great. Awesomeness like Postgresql's supporting every data type imaginable, Redshift and Aurora, file-based db's like SQLite...

So, yes, SQL won. But so did NoSQL. And all of the rest of us too.

collyw2• 11 years ago

The problem is people shoehorning relational problems into non relational databases. Then no one wins.

realusername• 11 years ago

I would say that people are too quick to jump on the new trend and are not evaluating what they actually need. In my company we are using PouchDB/CouchDb because of the exceptionally good sync engine, we have around 20 nodes offline at least 50% of the time, no SQL database could compete with PouchDB/CouchDb for this non-standard usecase, people should choose their DB according to what they need.

abannin• 11 years ago

SQL, as a language, will always win. There is no more elegant way to work with sets, and you can't work with data without thinking about sets.

I really like his point about the original breakthrough of SQL being the ability to abstract the way the data is stored from the application, and the language was just a natural outgrowth.

cloverich• 11 years ago

I'm curious about this stance and the proliferation of [IndexedDB](https://developer.mozilla.org/en-US/docs/Web/API/IndexedDB_A...).

I'm working on a simple note taking app, using IndexedDB and offline backups. Its been interesting working with IndexedDB -- but the primary problems I'm addressing are simple with SQL (Foreign Key's, join tables). I'm grateful to learn about dealing with the issue (and approaching it with very different solutions) with NoSQL -- but on the fence if I"m spending my time on something useful. I'd certainly be further on the path of working software if I could rely on something like SQL (via sqlite) being widely adopted in the browser.

bitL• 11 years ago

I am not sure how you can say "SQL won" - what I see is that client-accessible distributed hash tables have won (even PostgreSQL is now having them). For low-latency SQL is completely dead now.

woah• 11 years ago

Why are the concepts of a query language different than SQL, loose typing, flexible schemas, horizontal scalability, and a lack of joins all conflated as being the same thing- "nosql".

silon73• 11 years ago

To me, a major important thing about NoSql is the predictable behavior (because all queries are simple and there is no "query optimizer").

If I could ensure SQL never using index/table scans and make it fail without proper indexes it would be a major help.

ahachete3• 11 years ago

How not having a query optimizer makes queries more predictable? It makes them slower for sure, but not more predictable.

Indeed, I find NoSQL very unpredictable. When your query hits and index, the query goes well. When not, you usually end up doing a whole database scan. Plus you usually cannot use more than one index for a given query, and some NoSQL require an index just for ordering....

merb• 11 years ago

It doesn't matter where you store your data, as long as you always remember that the data you store is the value of your enterprise.

rjusher• 11 years ago

I do think it matters where and how you store your data, but I really like how he puts it simple, is not a battle between to opposing sides, simply there is a need for a data store that is not that structured, it is much simplier to prototype and achieve an mvp with a NoSQL Storage, but once you are in the real world you need to provide some type of guarantee to your users, in terms of data integrity.

In my opinion everybody needs to understand the differences and use each when it is the most convenient, use the best tool for the right job.

But maybe we will start to see some type of mixup between this two systems, because as he mentions in the article, OpenSource is driving the future of the databases, and due to this, the databases of the future would be more dynamic and more synchronized to the needs of the developer.

Database expert on why NoSQL mattered – and SQL still matters

Related Stories