From: <dea...@ve...> - 2003-09-04 18:06:58
|
Thanks Chris, That was exactly what I was looking for. I'm not going to be using a database, other than using a standard naming convention off the filesystem tree (solaris 8 ufs). I know using a database might improve performance, but the sheer amount of data we are dealing with really prohibits this esp. from a maintain-ability standpoint. Luckily, my employer has dedicated a significant amount of hardware for this task, including 2 SunV880's with NetApps and a GIGE backbone. I'll be using perl to generate the data points and to populate the filesystem tree, then using a scheduling servet and the new java 1.4 nio features to read-in the data and gen the graphs. Once we can get a proof-of-concept platform going, maybe we could use some optimization consultancy from the developers. Keep those opts rolling Nathaniel! -dean "A journey of a thousand miles must begin with a single step." Lao-tzu, The Way of Lao-tzu |