Core Spring Data
Written by: Oliver Gierke
Featured Refcardz: Top Refcardz:
  1. Git
  2. Prep for CD
  3. DNS
  4. Data Mining
  5. Spring Data
  1. Data Mining
  2. Spring Data
  3. DNS
  4. Machine Learning
  5. MongoDB

Link Details

Link 386733 thumbnail
User 251382 avatar

By chandru.in
via stu.mp
Published: Mar 28 2010 / 19:33

First, and foremost, I haven’t heard of anyone saying MySQL or PostgreSQL on comparable hardware is faster than NoSQL options. The best I’ve heard is that MS SQL setups on SSD drives with lots of RAM could do 6,100 result sets a second. I guess, based on these posts, I’d like to ask a few questions to the people who honestly think RDBMSs can compete with NoSQL solutions at large scale.
  • 33
  • 5
  • 6236
  • 11

Comments

Add your comment
User 289539 avatar

michele.mauro replied ago:

0 votes Vote down Vote up Reply

Why all these flames? NoSQL datastores cover use cases and types of data where the relational model (and then RDBMS) are not the best solution. So, why bother?
The relational db is not going away, for it has its own use cases that the NoSQL model can't cover.
It's just the usual "right tool for the right job"...

User 393686 avatar

RawThinkTank replied ago:

1 votes Vote down Vote up Reply

They are both used for different purposes, so there will be no wars, only those who hav no idea about where to use them will create such topics and waste time and emit carbon.

User 731965 avatar

GandalfSoftware replied ago:

0 votes Vote down Vote up Reply

Why not use a database that combines a key/value store, object database, relational database, and speed of an in-memory database? Take a look at InterSystems' Caché and Ensemble products...,

User 368023 avatar

yakkoh replied ago:

0 votes Vote down Vote up Reply

Another thing I don't like about RBMS and more specifically about the 'language' SQL is the mismatch between PHP (say) and SQL. Moreover I don't like SQL notation, the COBOL like syntax.
why not something like:
update tables=cars cols={wheels: 4, color:'red'};

User 289539 avatar

michele.mauro replied ago:

0 votes Vote down Vote up Reply

SQL comes from an era when the fastest way to bring 2 megabytes of data from Italy to France was by train, in a well padded (and somewhat heavy) suitcase. SQL is 36 years old, so it's not surprising that it has quite an "old" feeling. It has served us well, but now we are discovering some use cases that are better served by another repository model, this one just a little younger (Berkley DB is a keystore, after all, and it's 25-ish)...
These use cases happen to be quite hip and popular these days, so there's a lot of young people that takes everything old and wants to throw it away; but they forget that many old things, like 1 and 0 for example, or email, or TCP, are serving us really well, even now.

User 731965 avatar

GandalfSoftware replied ago:

0 votes Vote down Vote up Reply

True, but it really doesn't have to be an either/or situation... there is a lot of value is being able to use SQL to access your data... but also being able to access it as a multidimensional array... as an object database... or for that matter, being able to use the MultiValue/Pick query language (which has been around for almost 40 years). It is possible sometimes to have your cake and eat it too!

Add your comment


Html tags not supported. Reply is editable for 5 minutes. Use [code lang="java|ruby|sql|css|xml"][/code] to post code snippets.