What is Cloud Computing and How will it Affect Libraries?

As with many questions about the future of technology, I’m afraid that the only completely honest answer to the questions in the title of this post is probably “I don’t know and neither does anyone else”. So pardon me while I make a provisional attempt to answer this unanswerable question.

If you’ve used any of the popular Web 2.0 services over the past few years (e.g. Gmail, Wikipedia, Flickr or Twitter), you already have some experience with cloud computing, since most of these applications are hosted in the large online data centers that are the hallmark of cloud computing.

I think the utility analogy is helpful for someone who’s trying to get a handle on the nature and potential impact of cloud computing.  Like water and electricity, a computing cloud is a communally-shared resource that you lease on a metered basis, paying for as little or as much as you need, when you need it.  As this animation makes clear, you pay a limited amount when traffic to your web site is low, but the cloud responds quickly and increases the amount of bandwidth and other resources available to you as demand for your site increases. In other words, you aren’t paying for a basement full of servers waiting for the one day a month when your web traffic peaks.

As Charles Leadbeater points out, another helpful analogy is embedded in the phrase “cloud computing” itself. Real clouds are Photo by BasicGov; Creative Commons share-alike licensePhoto by BasicGov; Creative Commons share-alike licenseall made from the same fundamental component (i.e. water droplets), and these atmospheric clouds morph continuously from one type (e.g. cirrus) to another (e.g. stratus or cumulus). Similarly, shared online computing resources are the fundamental common component most modern networked applications and communities run on the same basic mix of IT infrastructure. The flexibility and scalability of cloud computing means that virtual clouds can form and dissipate as often as real clouds, depending on the interests and demands of end users.

Eventually, one of more of the following effects of cloud computing will probably impact libraries and other small-to-mid-sized organizations:

Cost savings

In an era of shrinking budgets, it gets harder with each passing year to justify the purchase and maintenance of servers that aren’t in use almost all the time. Cloud computing offers price savings due to economies of scale and the fact that you’re only paying for the resources you actually use.

Flexibility and innovation

Organizations of all sizes can take more risks when it comes to creative, innovative technology ideas when the new application will run on someone else’s infrastructure. Libraries don’t have to decide between devoting their limited server resources to the OPAC’s overflow traffic and a new mobile web application that one of your colleagues wants to develop. If they’re both hosted in the cloud, the resources devoted to each will shrink and expand as traffic rises and drops. Furthermore, creating and configuring new virtual server instances is fast and easy in the cloud. On the other hand, in this interview, Whitfield Diffie points out that in the long run the cloud might be more restricted and rule-bound than traditional IT. He compares the cloud to public transportation providers such as airlines which rely more on rules and fixed schedules than privately-owned planes.

Broad, general IT skills vs. deep, specialized skills

Cloud computing increases the pressure on IT professionals to become well-rounded employees with highly-developed managerial skills.  Knowing how to configure and network a server isn’t enough. Systems librarians have to manage complex projects and evaluate competing vendors on a variety of criteria. Holding vendors accountable is especially important when they manage a significant chunk of your online data and IT infrastructure. Therefore, as long as cloud security remains a significant concern, techies may be called upon to help write binding, enforceable contracts that hold vendors to certain standards with regards to reliability and security of their services. Furthermore, techies will likely be part of the teams that periodically audit cloud vendors and ensure they’re performing up to the contracted standards.

Cloud OPAC and Cloud ILS

Over the past year, more and more ILS vendors have started offering cloud-hosted versions of their products. As these posts from  Richard Wallis and Marshall Breeding explain, OCLC joined several other vendors last year when they began offering a cloud-based ILS tools that complement their existing cataloging tools (e.g. WorldCat and FirstSearch). As individuals and members of organizations, we’re already choosing between desktop applications and cloud applications when it comes to e-mail, RSS, file storage, word processing and other simple applications. Sooner or later we’ll have to make this choice for mission-critical enterprise applications too.

Private clouds, hybrid clouds and community clouds

Libraries may soon be building and managing their own data centers. In addition to all the hype and  optimism surrounding cloud computing, there are still significant fears and doubts Industry Challenges points out. In particular, the major cloud computing vendors haven’t yet fully addressed concerns about security, privacy and reliability. These concerns are leading some companies to build their own private or hybrid clouds. A hybrid cloud is primarily based in a privately-owned and operated data center, but it can shift some of its traffic and data processing requests to public cloud vendors such as Amazon or Rackspace on an as needed basis. This hybrid model would let libraries maintain more control over the applications and data stores that contain sensitive, private information about patrons.  Moreover, libraries can continually adjust and fine-tune the balance between the tight control of a private IT infrastructure, and the flexibility and savings of cloud-hosted infrastructure. If reliability or security of one vendor becomes a concern, you’re not committed to one company or one model of computing services. Moreover, if the thought of “building and managing your own data center” sends chills down your spine, you might consider a community cloud. As this GovTech article explains, Google plans to launch a government-only cloud this year to address government concerns about security and privacy. Just as libraries presently cooperate with one another to buy IT equipment, bandwidth and the services of IT professionals, libraries may soon cooperate in the building and management of data centers.  Alternately, if enough libraries express interest, a company such as Google, Amazon, Microsoft or another cloud vendor might create a Library Cloud similar to Google’s Government Cloud.  Or, a library vendor with deep IT resources (e.g. OCLC or SirsiDynix) might build library-centric cloud services on top of cloud infrastructure leased from one of the more established players.

Further Reading

For a high-level, big picture overview of cloud computing and its impact on culture, read Cloud Culture: The Promise and the Threat at Edge.org. This article builds off The Second Coming: A Manifesto a fascinating essay by David Gelertner on the near-term future of computing and digital culture. In Why Should Nonprofits Care about Cloud Computing? on the TechSoup blog, Anna Jaeger offers a nonprofit perspective, with a focus on the green benefits of cloud computing. Learn more about cloud computing for your nonprofit or library on TechSoup's cloud page. This page is a portal that aggregates a wide array of cloud-computing resources, especially those with relevance to nonprofits and libraries.

For more practical, technical explanations of cloud computing, check out

For a discussion of problems and concerns about the digital cloud, read: