Did you hear about how storms and extreme heat (in the eastern US) hampered Amazons Cloud Computing operations?

Did you hear about how storms and extreme heat (in the eastern US) hampered Amazon’s Cloud Computing operations? Amazon Cloud Computing operations

Despite recent outages, the cloud remains the best choice for businesses, here’s why

If you’re located in the US, you were most likely affected by the recent record-breaking heat wave and string of storms (unless you’re a lucky Californian). The phenomenon actually left millions without power; citizens and businesses alike suffered at the hands of this freak occurrence. Despite their well-preparedness, Amazon and their cloud computing operations were also affected by the high temperatures; their web services being knocked out temporarily. The Amazon outage put some major sites offline like Netflix, Pinterest and Instagram, bringing a firestorm of complaints down on these individual businesses and their subscribers, no doubt. Amazon Cloud Computing operations

But the real question is, aren’t we being a bit too harsh with Amazon and their cloud computing services? After all, Amazon’s Virginia data center was itself knocked offline due to a regional power failure. Shouldn’t the affected users and businesses be coming down on nature or the failure of the power companies to prepare for such an event? If you step back and look at the situation from a wider vantage point, all of the tech articles circulating the net proposing that Amazon and cloud computing are somehow weak or faulty seems utterly laughable. Keep in mind folks, this recent heat wave was completely unprecedented; thousands of temperature-related records fell. Instead of jumping on the Amazon hate-wagon, ask yourself this – is there any form of computing / networking solution which can hold up in such stringent situations?

Storage services for the aforementioned websites were lost on Friday and restored on Saturday; although everyone was collectively freaking out, this was by all accounts, a minor event. If you need further proof that Amazon’s cloud services are up to snuff, just take a gander at their track record; in the past 6 years they have consistently delivered continuous service without any major hang-ups or failures. Sure, there was the technical event in 2011 which brought Foursquare and Reddit offline, but these types of events are extremely rare.

A learning experience
The truth is that a little downtime resulting from technical glitches and weather-related anomalies should be expected. Cloud computing is still being perfected, with the dawn of every passing day comes new challenges and learning opportunities. If anything, these types of events are teaching us how to better prepare for the unexpected. All in all, it’s much better to face minor outages which provide us with true insight than to encounter “butt-busting” scenarios in the midst of years of perfect operation. Perhaps a good comparison would be the way the immune system works in the human body; the more practice it gets, the stronger it becomes. Conversely, those who live a life completely free of all contaminants often find themselves to be much more susceptible to diseases.

The tradeoff
The main reason that Amazon’s cloud computing services present a danger to sites and businesses is due to its centralized nature. This is to say that because you have a large amount of resources resigned to one location, the statistical probability of it causing outages for multiple businesses increases slightly. On the other hand, the alternative (traditional IT / Waterfall) only appears to be more stable because it is spread out over a larger area (non-centralized). In other words, while having a centralized computing solution presents (a remote possibility of) potential service failure issues, it cannot be argued that cloud computing is anything but superior in terms of what it actually delivers to customers. Virtually the only thing that traditional IT has going for it is it’s decentralized nature; but once again, it is nowhere near as powerful, efficient or useful as its cloud computing counterpart.

A more decentralized form of cloud computing?  
One potential solution for these types of disruption issues is to strategically place data centers across regions, nations, and continents. Perhaps as we encounter more and more breakdowns resulting from weather anomalies, malicious attacks and human error, a more decentralized, or shared approach to cloud computing will emerge? As more people and businesses adopt cloud computing, providers will undoubtedly begin building more data centers to handle increasing workloads. Once this occurs, data copying / sharing customer storage accounts across multiple centers is the next logical step to assure uninterrupted service.

Let’s not jump to conclusions here…
Once again, we shouldn’t immediately blame Amazon’s cloud computing infrastructure for service and power-related failures, especially when facing freak events such as the recent heat wave. I’m sure there were also a number of IT firms and cloud computing providers who were also experiencing major issues as well; singling out Amazon in the midst of this mess just seems silly.

Is your organization and IT department prepared for cloud computing adoption / transition?  
It’s been said thousands of times by hundreds of tech gurus and bloggers – cloud computing is going to replace traditional IT. If your organization is reliant on in-house IT services and/or web-delivered services / products, you are going to have to face up to the fact that cloud computing will affect you and your competition. But this isn’t a bad thing; through cloud computing you’ll have more options and possibilities at increasingly lower costs of operation.  What’s the best way to prepare for the inevitable cloud computing adoption wave, you ask? Affordable and technically thorough cloud computing e-learning is the way to go. Get your entire IT department trained and certified in one of many cloud computing disciplines today! Group rate discounts are available!