Sean Wallbridge
itgroove Recommendations from the Field
The following are recommendations from itgroove for Server (virtual or physical plant) Best Practices. Customers may not always listen or have the budget, but it won’t stop us from trying to protect them…
- Redundant Power Supplies – protect yourself against a single point of failure
- UPS – (Uninterruptable Power Supply) don’t count on your electric company to shut down your equipment ‘gently’
- RAID – (redundant disk arrays) and a hot spare drive
- Good Backups – system and application level software and offsite storage
- Leave Room for Scalability – Don’t max out the server day 1. Allow for additions/scaling, a server jammed with RAM or disk has no room to grow, rapidly to adjust to your needs
- Get the extended warranty – It is always far more expensive to pay for the downtime or recovery than it is to get the extra warranty
Downtime at an average of $60 salary per user per hour equates to a 100 person office burning $100 per minute when a critical server is down such as the domain controller. I am not sure what a modest priced new server without software would cost but for calculations sake, let’s say $6000 ( I am not a Network Admin so may be waaaaay off), which equates to 60 minutes of downtime in a 100 person office (Scheduled at lunchtime, not much problem!). Add to this the green energy savings of these new servers and you are likely talking 45 minutes of downtime which is cutting into the workday even if you can schedule your server failures. Hmm, if I were looking after the company budget, I don’t think I could not follow your recommendations.
thanks Sean.