a central point for collection of information as it relates to cloud computing in the government
September 12, 2009 at 7:42 pm #80280
A little different perspective, but provides good overview of “the cloud“.
Believe because of the audience of this article, that it COULD go a long in PERHAPS making the technology understandable by a larger audience.
From the Library Journal
A Guide to Cloud Computing and Storage
Author: Ellyssa Kroski
The past few years have seen the phrase ‘cloud computing’ steadily gaining momentum, becoming one of the most hyped technology buzzwords since “Web 2.0”—and with nearly as many different definitions.
But one thing that most can agree upon is that cloud computing is a major shift in the way we’re approaching computing as both individuals and organizations, and is being referred to by many as an IT revolution. Gartner’s 2009 Hype Cycle for Emerging Technologies has cloud computing high on the list among technologies that it predicts may be transformative to the IT industry in the next two to five years.
What is it?
Simply put, cloud computing means using Web services for our computing needs which could include using software applications, storing data, accessing computing power, or using a platform to build applications.
Through Web 2.0 applications, we have already been shifting much of our personal computer usage to the cloud. Millions of us have freed our bookmarks from the desktop by storing them on social bookmarking websites such as Delicious, uploaded and shared our videos on YouTube, used services such as Slideshare to host our presentations, worked collaboratively on Google Docs, and adopted web-based email services such as Gmail—some of us may even have designed applications for popular platforms such as Facebook. According to a September 2008 Pew Internet & American Life study, 69% of online Americans have already participated in cloud computing activities.
But cloud computing is about more than just these types of consumer-level activities—from a business perspective it can be a viable, cost-effective way for organizations to run or complement their IT systems. There are three major types of cloud services available:
Software as a Service (SaaS)
Applications or software is delivered as a service to the customer who can access the program from any online device. Some of these Web-based applications are free such as Hotmail, Google Apps, Skype, and many 2.0 applications, while most business-oriented SaaS, such as SalesForce, is leased on a subscription basis. There is usually little customization or control available with these applications. However, subscribers benefit from low initial costs, have access to (usually 24/7) support services, and needn’t worry about hosting, installing, upgrading, or maintaining the software.
Platform as a Service (PaaS)
With PaaS, a computing platform is provided which supplies tools and a development environment to help companies build, test, and deploy Web-based applications. Businesses don’t need to invest in the infrastructure required for building Web and mobile applications but can rent the use of platforms such as Windows Azure, Google AppEngine, and Force.com. Applications which are built using these provider’s services, however, are usually locked into that one platform.
Infrastructure as a Service (IaaS)
This type of cloud computing is also sometimes referred to as HaaS or Hardware as a Service and it involves both storage services and computing power. Amazon’s Web Services, one of the major players in this area, offers two main products including the Elastic Compute Cloud (EC2), which provides computing resources, and Simple Storage Service (S3) for data storage.
Companies are using Amazon’s Web Services to host or backup their websites, for content delivery, to run high performance computing simulations, to host media collections, and much more. Most of these cloud services are available on a pay-per-usage basis, differing from the SaaS subscription model, enabling customers to scale up or down depending on their need at any given time and only pay for what they’ve used.
Let the cloud do the lifting
This scaling can provide significant cost savings and the option of sidestepping major capital investment. Instead of spending on the purchase and maintenance of servers, software, and infrastructure, they are rented. Companies are able to scale instantly and on-demand, paying only for the processing power, bandwidth, or storage space they use.
If using a major provider, there is low risk involved for downtime or data loss as these companies usually have redundancies and backups in place, as well as tested security measures. Cloud services make data and computing capabilities portable, sharable, and accessible from any online device, are OS-neutral, and usually easy to use.
Businesses are finding that by outsourcing some of their IT heavy-lifting—such as website hosting—to the cloud, they have more time and resources to dedicate to new projects. Managing VP of Gartner, Daryl Plummer, estimates that eight out of ten technology budget dollars in corporations go toward maintaining systems rather than innovating.
Eggs still aren’t safe in one basket
There are definite trust issues involved with working in the cloud, however. Data and systems “live” somewhere else, beyond the safety of an organization’s firewall. This raises a major concern about the provider’s reliability—after all if their systems go down, so do all of the websites, data, and potentially business-critical applications they are hosting.
While major cloud services providers have generally exhibited a great deal of stability, they have also had some hiccups in service. At the ALA Chicago LITA Top Tech Trends 2009 panel, Clifford Lynch, director of the Coalition for Networked Informaton (CNI), strongly urged caution, saying that a large-scale failure of cloud services is inevitable. “I guarantee you, there will be [a failure] in due course,” he said, one that will force a recalibration of the faith people put cloud services for critical data preservation.
Having data housed elsewhere may also bring up security and compliance issues with some types of information and legal issues with others.
The question of data ownership has been a key topic of discussion lately with regard to trusting in the cloud, especially after Facebook’s move to revise its TOS to include perpetual ownership and right to use any and all content stored or posted to its site. And Amazon did little to gain user’s trust with an ironically Orwellian intrusion of owners’ Kindle devices in order to delete copies of 1984 and Animal Farm.
Libraries and the cloud
Libraries will want to consider what types of information or processes they want to trust to the cloud. Eric Lease Morgan also spoke about the responsibility of libraries to preserve information at theTop Tech Trends panel, making the point that outsourcing its preservation in effect relinquishes that obligation.
Libraries will need to consider not only this type of ethical quandary, but also practical ones such as the privacy of sensitive information such as patron records, and concerns about records retention requirements. But it needn’t be an all-or-nothing decision as libraries may choose to continue to host some of their own systems while using the cloud for less sensitive processes such as hosting library websites, backing up media collections, or storing and accessing bibliographic data.
Libraries have already begun to adopt cloud services to alleviate their IT departments and increase efficiency.
The OhioLINK library consortium is using the Amazon’s Web Services to host a handful of their Digital Resource Commons repository instances such as Kent State’s Centennial Collection, and is testing server administration in the cloud, as well as the limits of DSpace repository software.
The District of Columbia Public Library is using Amazon’s EC2 service to host their website and according to Director of Information Technology, Chris Tonjes, “This provides us with rapid scalability, [and] redundancy (if one Amazon data center fails, a mirrored version of our site is on another and can come online in less than 30 minutes).”
They are also using Amazon’s S3 service to backup their ILS, their upcoming digital repository uses Flickr, and Amazon EC2, and the District of Columbia government will soon be adopting the GAPE version of Gmail as its enterprise mail platform. Tonjes adds that “Server vitualization + cloud computing give us great speed and flexibility and typically are significantly cheaper than enterprise data center-based infrastructure.”
The Eastern Kentucky University Library is using Google Docs to collect responses to web forms, Google Calendar for instruction and meeting rooms, and Google Analytics to collect statistics about their website, catalog and blogs.
At Western State College in Gunnison, Colorado, they are using Google’s App Engine for their ELibrary, and have also migrated two Microsoft Access databases that they used for serials circulations and government publication management to that service.
In addition to these libraries, the Library of Congress has entered into a partnership with Duracloud for a one-year pilot program testing out cloud storage capabilities (see last month’s interview with LC staffers for more), and OCLC has announced a new Web-scale, cooperative library management service.
Big or small, transition is under way
Nicholas Carr likens cloud computing to the transition from companies supplying their own electrical power to plugging into a centralized electric grid in his book The Big Switch, while Simon Wardley makes comparisons to the industrial revolution at the 2009 OSCON conference. Whether you agree that this style of computing is as revolutionary as advocates profess, it is certainly an important shift to be aware of, especially as libraries and organizations struggle with dwindling budgets and staff.
©2009 Reed Business Information
You must be logged in to reply to this topic.