Five Deadly Sins of Disaster Recovery Planning
Common blunders that result in data recovery disasters.
The quality of a network is determined by a number of things, including cost, reliability, reach and the level of non-emergency quality of service.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=i
The most glaring and dramatic of issues are network outages. Though there are some high-profile cases, the track record of the major carriers actually seems pretty solid. Viscerally, it doesn't feel that way, since we hear a lot about outages. But the reality is that we never see a news story or read a blog post that says that network X or Y is performing well and hasn't experienced an outage.
That may be so, there certainly have been outages during the past few months. TechCrunch and other sites report on a major outage on Verizon Wireless' 4G network on Dec. 6 and 7. The outage seems to have been inconsistent and extended down to Verizon services below 4G. AT&T suffered a similar-sounding outage last month.
The highest profile outage of the past few months no doubt was the one suffered by BlackBerry in October. Not only was it widespread, but it fit the BlackBerry narrative: A company that once had been great - indeed, revolutionary - had been overtaken by young upstarts and now was flailing and hapless.
Sure, eventually everything fails-anyone who's ever worked in a data center knows that. And the big cloud and managed service companies like RIM Amazon, Facebook, Google and Microsoft have built their infrastructures around the idea that things fail regularly, and have gone to pains to build resiliency into their systems. So why, if they're designed to deal with little failures-like a dying server, a lost drive, or a cut power line-do they seem to fail so often on a huge scale?
There actually are two very similar but subtly different issues here. The first is why there seem to be so many failures, and the other is why these failures seemed to be caused by a small flaw. It can be argued - I am not necessarily doing so, but just pointing out the fact - that based on the complexity of what happens on a wireless network, there actually aren't too many outages.
Put another way: One hears about accident after accident on a metro area rush hour traffic report. But, considering the number of vehicles traversing the roadways - and the inattentiveness of many of the drivers - it could be argued that there are surprisingly few mishaps.
The other question also is a matter of perspective. There are millions of elements in a network. While it is possible to back up the big ones, the small fries likely are on their own. Many probably fail, but only a select few conk out at precisely the wrong time.
The bottom line is that the reflective idea that wireless networks are unreliable may be true, but - contrary to common wisdom - it may not be.