That is not a question that can be answered directly... it come down to exactly how much data you expect to be handling, how many nodes you plan on using and what your proposed node configuration might be.. generally, a lot of RAM always helps and in a RAM-based solution like NDB, of course it's likely to come in handy, but if your data is small, adding RAM will have no effect. As for the effectiveness of multi-cores, again, what kind of concurrency load are you expecting?
The question is a fairly complex one and has everything to do with ones particular circumstances: that is why there are no simple answers. On 8/31/07, Sid Lane <[EMAIL PROTECTED]> wrote: > all, > > I am working on a budget proposal for next year to put in a MySQL cluster > but wanted to validate (or correct) a couple of assumptions: > > 1. do storage nodes benefit far more from additional RAM than they do from > faster CPUs/multiple cores? > > 2. do SQL nodes benefit more from faster CPUs/multiple cores? > > basically, I am trying to find the best bang/buck configuration for each > server type and am assuming it's better to spend more $ on RAM for storage > nodes and CPU for SQL nodes - is this correct? > > is anyone aware of any benchmarking that has been done with various > configurations? > > any help appreciated... > > thanks! > -- - michael dykman - [EMAIL PROTECTED] - All models are wrong. Some models are useful. -- - michael dykman - [EMAIL PROTECTED] - All models are wrong. Some models are useful. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
