Vinayak Mahadevan <[EMAIL PROTECTED]> wrote on 27/09/2005 11:28:51: > [EMAIL PROTECTED] wrote: > > >Vinayak Mahadevan <[EMAIL PROTECTED]> wrote on 27/09/2005 04:55:13: > > > > > > > >>I am creating an application in Visual Basic 6.0 which will require a > >>centralised database server. All this while I had been planning to use
> >>MS-Access. But then I found out that MS-Access is ok to be a desktop > >>rdbms but not for an enterprise level rdbms. So I am planning to use > >>MySQL as the backend for the application. What should be the minimum > >>system requirement to run the database on. > >> > >> > > > >MySQL can run on almost nothing. The question is not what system you need, > >but what performance you want. I think you *could* run MySQL on a P200, > >Win 98, 128Mb ram, 40Mb disc. But the performance you would get would be > >seriously disappointing. > > > >You need to think what size of database you want, how many queries and > >updates per second you will need, and how complex your queries will be. > > > > > > > > > > > > > > > > > Thanks for the response > I am planning to run it on a x205 series IBM Server which right now has > 256 mb ram but will be upgraded to 1 gb. And the maximum number of > connections at any point of time will be say around 10 That sounds reasonably competent hardware. But it is not the number of connections that matters, it is the number and complexity of queries. One connection can generate a massive query which will lock out others; idle connections consume a small amount of memory but no other resources. I can only suggest you try to set up a representative test load and see if the performance is adequate for you. If performance is not adequate, after having examined your slow queries carefully and checked your indexing, extra ram is the first hardware upgrade to do. However, I think newcomers to MySQL are frequently surprised by its performance once properly indexed; don't spend money on extra ram until you have tried out a real (or simulated) test. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
