Pantek Library
Hosting Provided By
CybrHost
High Speed Hosting

Re: cluster takes enormous room to store data

From: Gaël Marmillod <le.peru.vient(at)gmail.com>
Date: Mon Jul 09 2007 - 08:52:14 EDT


Hello everyone,

Firstly thanks a lot to all of you for your answers and hints.

So, if I get it all:

  • as long as I am on 32bit machines, so of course 32bit OS and MySQL
  • as I'm running low on time
  • ...no means to raise the memory used except through not-so-recommended (according to documentation) solutions like running multiple nodes on a same server. (anyhow, this solution would probably solve nothing, just get me stuck a little further in the data import)
  • ....no time and maybe no possibility to rearrange data types, schema and such

My best way out seems to evolve to 5.1, to spare on Data Memory consumption, right ?

I'll surely have a look at this ndb_size.pl tool, it'll help to balance between Data and Index Memory, but is the way of storing so different that it can take ten times the space it took previously ? I don't know the DB I try to import in details, but there are indeed a lot of TEXT and VARCHAR in them. And again, I'm no expert, so if anyone tells me that for example, every TEXT, even thoug weighing a few bytes, takes minimum 5k, I'll take his word for it ;-)

Thanks,

Gaël Received on Mon Jul 9 08:52:38 2007

This archive was generated by hypermail 2.1.8 : Mon Jul 16 2007 - 05:11:01 EDT

Do you need help?X

Contact Us  Legal Notices  Order Services Online 
Pantek Home  Privacy Policy  IT news  Site Map  Pantek Library