mike, You should look at real-world examples sometime.
In one lab where I spend a bit of time, there are about 80 computer server systems. Five years ago, all of them had RAID, and there would be a hard drive failure out of the 300 or so hard drives at least once a month. Pull it out, drop a new one in. About three times a year, a RAID controller would go bad - disassemble everything, put in the spare identical RAID controller, be back and running in a day. Every year, maybe twice a year, there would be a RAID controller failure and no more replacements. Then you rebuild everything from backup onto new hardware. Today, there are some RAID on some of the older servers, but no one buys a system with RAID any more. Not worth the expense of having to buy multiple, identical RAID controllers at the same time. And, the current best practices are often to use shared storage on the network - NAS, SAN, etc. Some of the NAS have software-based RAID mirroring to allow for a drive failure without losing data for even a minute. But what the admins are really looking at is having the same data replicated on multiple NAS. You can do a lot of that with a good SAN, but interoperability issues and expense can limit you there. Also, hardly any server runs its own local database. There are dedicated database machines, some with Oracle, some with MySQL, and they are all replicated to other machines on the network. Does this help you any? On Sun, Sep 7, 2008 at 6:59 PM, mike <[EMAIL PROTECTED]> wrote: > I'm still waiting for the > solution that improves on the old RAID technology..now we sit and wait for > -- John DeCarlo, My Views Are My Own ************************************************************************* ** List info, subscription management, list rules, archives, privacy ** ** policy, calmness, a member map, and more at http://www.cguys.org/ ** *************************************************************************
