up vote -2 down vote favorite
share [g+] share [fb]

Major use of that cluster would be to have add/delete/edit operations on individual records based on primary key.

Database itself is not complicated but main concern is handling the operations on that amount of data efficiently.

And database would be mysql and what amount of RAM would you suggest?

link|improve this question
1  
This question is way too vague. Trillion can be either 10^12 or 10^18 for starters. – Iain♦ Jul 19 '11 at 16:19
its 10^12 why it should be 10^18? :o – Gary Lindahl Jul 19 '11 at 16:40
ah sorry for that confusion.. I meant 10^12. – Gary Lindahl Jul 19 '11 at 16:49
1  
One hundred billion dollars! </drevil> – Tom O'Connor Jul 19 '11 at 16:49
feedback

closed as not constructive by Iain♦, pauska, Scott Pack, sysadmin1138♦ Jul 19 '11 at 16:38

This question is not a good fit to our Q&A format. We expect answers to generally involve facts, references, or specific expertise; this question will likely solicit opinion, debate, arguments, polling, or extended discussion. See the FAQ for guidance on how to improve it.

1 Answer

up vote 2 down vote

And database would be mysql

Mysql is seriously inadequate for the task.

Get Db2. On mainfarmes. Call IBM. I mean LARGE mainframe.

NOTHING scales like DB2.

ANY idea what you talk her?

what amount of RAM would you suggest?

What amount of DISC space? What amount of IOPS you need?

Hm, maybe call oracle. Their EXADATA stuff should scale up like that - cost is some millions for that.

RAM is not your problem, the rest is, or you cluster into thousands of nodes.

link|improve this answer
feedback

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