Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Possible Duplicate:
Can you help me with my software licensing question?

we have a ms-sql 2008 standard edition. The db became too huge, about 8 10^9 records.the db files are about 4.5tb each. We cannot effort us to get enterprise edition to slice the database. We need partitioning.

So the idea is to use Mysql cluster with many datanodes. We already started to move data.

I wondered do we need to buy a licens for mysqlcluster?are there performance difference between community edition and commercial one?

Thanks Arman.

share|improve this question

migrated from stackoverflow.com Mar 12 '11 at 23:13

This question came from our site for professional and enthusiast programmers.

marked as duplicate by Mark Henderson Jan 22 '12 at 21:59

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

2 Answers 2

The business has generated that much data, but they won't fork over the money to keep the application running? I'm sure that converting everything from Microsoft SQL to MySQL will probably cost more than the SQL Server Enterprise Edition license will cost you.

share|improve this answer
    
Hi,thanks for answer. Actualhis is a research project,not a buisness, we try to accelerate the data processing pipeline. –  Arman Mar 13 '11 at 9:59

First off, I know very little about MySQL, so I can't speak to that, but I suspect migration is probably not the best plan.

In SQL Server, there are more options available than going straight to table-level partitioning, which is Enterprise and Datacenter Editions only:

  • Per your comment on the other answer, if this is simply a research project, change editions to Developer Edition -- this is a Datacenter engine with a licence that doesn't allow you to expose the instance in production. This may or may not be suitable. Cost is ~$50 for this edition, and you can do all the table-level partitioning you want.

  • Data can be partitioned using partitioned views with multiple backing tables (available in all editions). This gives you the advantage of partition elimination at the cost of manageability.

  • Data can be split over multiple files in a single file group and/or each table can go in its own file group (available in all editions). With terabyte amounts of data, you should be doing this anyway, but I'm mentioning it in case you're not.

What's interesting is that these techniques can be used in any combination. For pure performance, consider implementing all of them.

The best strategy, however, is unclear without more information about (a) the structure of the data, and (b) what you're trying to do with the data.

share|improve this answer

Not the answer you're looking for? Browse other questions tagged or ask your own question.