"My data’s now stored in 'the cloud,’ right?"
"Yes - but do you really understand what that means?"
"Yes. ... No ... It’s 'up there’ some place, right? Is 'the cloud’ a real place or is it an imaginary one?"
.. and therein lies the tale.
Computers process data and turn it into information. They must store the data/information that they process/create someplace. One of the first technological leaps with large computer systems was changing the method of input from punch cards to keyboard terminals. We called the large computers mainframes, and they stored the data on magnetic tape, large disks and drums. Users used the keyboard terminals for input and to view and analyze the data.
When personal computers arrived in the late 1970s and early '80s, they acted as little mainframes, doing all the processing and storing of data locally. They first used cassette tape as a storage medium, then removable floppy diskettes, which held between 140,000 and 320,000 characters. Finally, large hard disk drives arrived, and grew from early small capacities of around 1 million characters (10 MB) through to many billions of characters (500 GB) to multiple trillions of characters (2 TB). Storage has gotten bigger in capacity, smaller in physical size, and much, much cheaper.
Yet even with the breakthroughs in storage cost, capacity and size, there are still issues. We needed to share data with others; that led to networking and file servers, very high capacity disks that could be shared by groups. Businesses have dealt with these problems and today, they often use mainframes as their central servers.
What has, however, become a recent phenomena is multiple devices (desktop computer, laptop, tablet, smartphone) and users' desire to access data from all of their devices from anywhere. When there were only desktops and laptops, a user could carry around USB drives with reasonable certainty that they could be plugged into any computer and the information used.
There were, however, other approaches. One of the early services to store information on the Web was Hotmail, which was, at first, an independent operation and was then acquired by Microsoft. This service allowed users to keep their email processing online, rather than relying on programs such as Outlook or Eudora to bring mail from servers to local PCs. The Web-based service provided space for the storage of mail as well as mail-processing tools - and it was free. Yahoo Mail soon followed and, eventually, Google’s Gmail.
Announcement: We Have Moved to the CloudYahoo added chat facilities and space to store photos. Other similar services emerges. And most of us didn’t stop to consider just where our mail actually was or where we were chatting. Without even knowing it, we had moved into the cloud! (Learn more about what that means in The 5 Ways Cloud Technology Will Change the IT Landscape.)
Google soon added other functionalities to its services, lumping word processing and spreadsheets (and later presentation software) under the umbrella of Google Docs (now Google Drive). The advent of smartphones and tablets added some urgency to the cloud movement, because these devices didn't provide a lot of options in terms of moving data. Apple’s iCloud, introduced in 2011, added elegance to the process automation and automatic uploading of predetermined files. Amazon entered the fray even earlier, starting its own cloud service in 2002. Even more recently, DropBox gained significant market share at a rapid pace.
A user could use any of these services at low or no cost. All of a sudden, we were all in the cloud, a fuzzy amorphous place that held our data in some unworldly digital corral - at least that’s how it is portrayed and how it feels to most of us.
The reality is that our data is stored on servers in massive data centers throughout the country, data centers maintained by Microsoft, Apple, Amazon, Google, and many others.
Where Things Get CloudyWhen we hear about the cloud, what we hear most are about its promise. It provides better interconnectivity and access, it's often less expensive for businesses and it requires much less hardware. But there are a few dark clouds on cloud computing's bright horizon too. The New York Times recently ran a two-part series pointing out environmental problems caused by the humongous data centers that make the cloud work. Writer James Glanz points to large-scale - and often wasteful - energy consumption and air pollution.
Of course, as pointed out in an InformationWeek rebuttal article by Charles Babcock, many of these difficulties are eliminated in new data centers with state-of-the-art energy management systems and more judicious use of diesel back-up power systems. Even so, this is not a problem that's been entirely resolved in all data centers.
For example, when Microsoft purchased a 75-acre site in Quincy, Washington, for a data center in 2006, the community saw it as a boon to the area, at least at first. But the bloom soon came off the rose and, as Glanz tells it, "the gee-whiz factor of such a prominent, high-tech neighbor wore off quickly." First, the community tackled the company about 40 giant diesel generators at the facility, which Microsoft had installed for backup power. Community members worried about their proximity to an elementary school.
Then, Microsoft went head-to-head with the local utility provider by proceeding to waste millions of watts of electricity in an effort to erase a $210,000 penalty it owed for overestimating its power use.
A Microsoft spokeswoman said the episode was "a one-time event that was quickly resolved," but the problems reveal a tug-of-war that's likely to continue as data centers get bigger and appear more place across the country.