It's been a difficult Easter for online computing services: The Amazon EC2 service failed and Sony's Playstation network was out of service over the important holiday break and continues to be down as I write this.Much has been written about these failures already, with plenty of debate particularly from those who doubt cloud computing's credibility.
It's been a difficult Easter for online computing services: The Amazon EC2 service failed and Sony's Playstation network was out of service over the important holiday break and continues to be down as I write this.
Much has been written about these failures already, with plenty of debate particularly from those who doubt cloud computing's credibility. The reality, as anyone who has worked in a large enterprise knows, is that computer systems go down regardless of their architecture.
We expect 99.9 uptime guarantees in a service driven society, whether from on-premise servers, new fangled cloud services or our home cable service.
IT professionals are all often the unsung heros of the enterprise, keeping systems up and running in difficult circumstances. Being prepared for utility computing outages and downtime requires different approaches for end users who are relying on the data in systems.
In the case of Sony's online Easter Playstation solstice, the thwarted gamer can go outside and get some sunshine, saving countless virtual digital lives from slaughter. For enterprise computing business users the stakes are a lot higher.
We accept there will be power outages of our home and work utilities - electricity, water and gas supply failing due to natural causes, accidents and breakages. Planning for these is Boy Scouts stuff - keep a stock of candles, water, canned food and a camping stove for worst case scenarios.
(That's Robert Baden - Powell, founder of the Scouts movement in the video clip above, who created the international scouting movement's motto 'be prepared' a hundred years ago).
Natural emergencies such as an earthquake are likely to see all services down for at least 72 hours, yet we have naively grown to expect our online shopping malls and services to be always there.
In the enterprise we are typically reliant on a variety of systems to keep the business ticking over and information flowing. Strategies for survival in the event of something being a problem are as old as running out of quill ink or jamming the photo copier.
Systemic failure of an email system or other critical communication systems can certainly slow things down but are largely manageable by using telephony and other ways of communicating. According to the National Academy of Sciences a major solar storm would shut down US power grids and fry satellites, leaving essential systems knocked out for months. This is a worst case scenario, but the next peak of this type of activity is expected very soon.
Planning for disaster survival is as important digitally as it is physically. Perhaps we need 'be prepared' guides for business information to keep things running like the excellent 72hours.org, with information on backing up crucial data and keeping it somewhere safe and other obvious but sometimes overlooked common sense ways of being prepared.
Anyone who has lost a hard drive of critical information knows how important it is to regularly back up all your stuff - there can be a sense of complacency in the business world that responsibility for looking after your work information is someone else's problem. This service economy mentality needs to be tempered with common sense around keeping track of your essentials and backing up where necessary.