As all Brits know, predicting the weather is not simple. Presaging where lightning is going to land is something of an impossible task.
Amazon will be fully aware of those two truths, the latter offering it a handy excuse for the outage caused by a lightning strike in Dublin on Sunday. Yet for some, that excuse won’t be enough to ease worries over the reliability of cloud computing.
It has not been long since the provider’s last outage, which took out a number of high profile websites such as Foursquare, Reddit and Quora for a short period of time. Given the proximity of these two events, it comes as no surprise some are feeling a little jittery about the prospect of staying in, or moving to, the cloud.
For the thousands upon thousands of companies who rely on constant internet presence, the affect of an outage can have serious negative bottom line consequences. Do instances such as the Amazon lightning strike therefore indicate the public cloud simply isn’t a viable option just yet?
The damage done
There is little doubt some customers’ faith in the public cloud has been challenged by the outage, which also took out Microsoft BPOS services.
The most unsurprising, and indeed most valid frustration vented by customers and industry pros has been over the lack of greater resiliency on behalf of the provider. People have rightly asked how just one incident could have taken down so many users’ services for so long. The overarching feeling is the providers should have been ready, especially in light of the outages seen earlier this year.
“It’s fair to say that accidents happen. It’s equally fair to say they are caused. There’s no question Amazon should have been better prepared,” said Phil Hambly, director and InTechnology.
“It’s a fact of life that technology breaks and those delivering a service have an obligation to underpin service guarantees. It’s about designing and building a properly resilient infrastructure, an absolute commitment to 24/7 monitoring and support, and a rolling commitment to maintenance and upgrades.”
For companies who base much of their business on Amazon services, the effects of the outage will have been even more pernicious. They will be the ones whose faith has been weakened the most.
“High profile cloud-based outages are enough to put any customer off from putting their confidence in a cloud computing supplier,” said Keith Bates, chairman of the Cloud Computing Centre. “After all, what company could survive without access to business-critical data for over 24 hours?”
Amazon said the outage was due to the back-up generators being taken out by the lightning. Normally, these back-up power supplies would automatically be brought into action if the main generator stopped operating. For Bates, Amazon’s excuse was not enough.
“It is of course possible for the national grid within a geographical area to collapse, and whilst it therefore may well have been a freak accident, Amazon and Microsoft should have been better prepared. Indeed, all data centres should cater for the possibility of losing power from a national grid,” he continued.
“In the event of momentary losses of power, Uninterrupted Power Supplies (UPSs) kick in as soon as power to the data centre cuts out, but a data centre cannot run off UPSs for long. Instead, if the power is suspected to be out for a considerable length of time, generators will kick in, but cloud providers need to guarantee with their fuel suppliers that fuel will be supplied until the situation has been resolved – had Amazon and Microsoft not prepared for this situation?”
Safe havens?
For a portion of customers, the bolt out of the blue may signal a retreat from the public cloud. In such cases, moving to a private cloud model, whilst costing more in initial investment and ongoing maintenance, can offer the succour companies feel they need.
“For public cloud customers, they are unable to examine the infrastructure on which their business applications will be running, nor can they be exactly sure where their data will be located,” said Bates. “At the opposite end of the spectrum, customers of private cloud solutions can be shown exactly where their sensitive data is to be held, to the extent of being given a tour of the premises of the data centre, so that they can be confident that their supplier is hosting their data where they want it to be hosted.”
Private may be enticing, but the hybrid model has been cited as the ideal option for the average business. Keeping the most valuable assets in house gives companies that additional peace of mind and extra control over where data is stored.
Moving everything to the public cloud evidently carries more risk than other options and hybrid gives businesses’ the benefits of cloud as well as any control they may need. Providers such as Citrix and VMware have been busy designing products to let businesses choose where they want to move different portions of their infrastructure.
For instance, Citrix’s Cloud Bridge offering announced at the company’s Synergy event in May is designed to do just that. Vendors and customers alike are looking at this model as the most sensible right now.
Waking up in the clouds
Yet despite the anxieties around public cloud, and the evident need for Amazon and Microsoft to perform better in ensuring back-up services themselves have back-ups (and those have back-ups too, and so on, and so on), the industry’s reputation should not be harmed by this week’s downtime. The cost and scalability benefits of the model will not be forgotten.
Make no mistake, this won’t be the last significant outage in the coming months and years, but people shouldn’t fear the public cloud more because of this Sunday’s blip. Somewhat paradoxically, they might want to fear it less.
For firms who want to continue to use the public cloud, the Amazon event will only act as a wake up call. Let’s not forget the public cloud industry is still a fairly nascent one, which has many lessons to learn. Amazon, as the leader in the market, will make the necessary changes. Furthermore, in making mistakes, it will show other providers how to avoid them and the ignominy which is intrinsically linked with outages.
According to Hambly, outages will lead to greater awareness about what cloud offerings are business-ready. “High-profile outages will help more than hurt,” he told Cloud Pro. “The Cloud is here. You can’t put it back in the box... Its reputation won’t be hurt but the distinction between consumer and enterprise grade is being established, and that’s to everyone’s benefit.”
As Hambly indicated, identifying the wheat from the chaff will be made easier as the market develops and outages occur. If companies feel vendors are not living up to their SLAs, they should look to other providers, not ditch the cloud altogether. If Amazon continues to fail, there are other options.
Furthermore, those who say keeping everything in-house is safer in the physical respect are, quite simply, wrong. Not one place on the face of the planet is safe from freak occurrences such as a lightning strike.
“Organisations must be reminded that both public cloud and private cloud solutions provide a far higher chance of uptime than an on-premise alternative. If a business was to opt for an in-house IT system, whereby all the power came from one source, operating on one single site, who knows what the potential downtime might be?” Bates added.
You can’t predict the weather. Any organisation’s servers could be in Zeus’s crosshairs soon. The clouds are just as safe as anywhere else, possibly more so.
