Originally Posted by PCCstudent
We are getting into all this analysis about using the least amount of resources and what the pros and cons of how to implement the recovery process,including making the best use of employee time.When all of these variables are pulled into the question "what is the best backup policy/procedure for company ABC" it seems like you can talk all night about the issue but still end up with a compromise.Thanks for making me feel that learning every theory of backups that exist is not really required.I would at least like to be able to get comfortable with some main techniques.Our instrusctor suggests a hybrid approach combining rsync with dervish,lots of reading to do.I feel backups is such a large area that trying to cover it in 2, 4 hour sessions only confuses people (well at least it confuses me).
your right this is something that can be discussed forever so just a quick overview.. In any commercial back up plan the first thing that comes to mind is Risk Management and Security, so here you place emphasis on first securing data,.. IMO this is first done at the network level with tiered HW firewalls, followed by "risk" what is sensitive I.E financial records, personal information etc so what data should be encrypted and what data can be left as unencrypted so ask, what is more at risk?
Back ups, now i assume we are talking huge amounts of data and enterprise level, not just your some home user.. here comes in Business Continuity planning, from my understanding these are policies that plan for data corruption, loss and even theft/viruses hence why back up.. so the main concern at this part is recovering from data loss/corrupted ASAP and maintaining business continuity.. any down time means loss of revenue or worst in the case of data theft.. back to business continuity, RAID is one tier to recover from data corruption, redundant storage servers that can come online as soon as a server goes down. backing up to offsite datacenter's offers another tier of data protection as in the worst case scenario you data can be restored in the case of total failure or environmental damages to your own servers such as fire.. this could go on for ever LOL but in the end it all comes down to business continuity planning, so for this to be effective data recovery solutions must be put into place and polices drawn up (disaster recovery)
IMO in an a large enterprise, no data should be stored on a users machine, thats why SANs and enterprise NAS servers were invented.. realistically only the OS should be installed on a users machine and apps in a repository of the company and even then you could run a zero client where even the OS is loaded from a server across the network Edited by stubass - 4/2/12 at 2:37am