Besides, a company run by many vs. one run by one isn't necessarily a panacea to availability problems -- in fact, I'm more comfortable with the current arrangement as bigger operators are more susceptible to change according to profit margins. Despite Tarsnap's great convenience, relying on a single backup point for mission-critical work is a recipe for disaster no matter how much you sugarcoat it.
-- M Preddy [email protected] On Mon, Mar 10, 2014, at 08:31 AM, Scott Wheeler wrote: > > On Mar 10, 2014, at 3:40 PM, Daniel Wilson <[email protected]> > wrote: > > > As I understand it, the underlying infrastructure is AWS ... so ... massive > > availability is already provided. It would be one of the most difficult > > systems in the world to bring down with an uncoordinated DDOS. > > AWS doesn’t make things scale with magic pixie scaling dust. You still > have to set things up specifically for auto-scaling to happen; I would > wager that the majority of sites / web-services on AWS *do not* do that. > > -Scott > > -- > Scott Wheeler | Co-founder | Directed Edge | www.directededge.com | > @directededge >
