This post sponsored by the Enterprise CIO Forum and HP.
Cloud storage appeals to enterprises as a less-expensive alternative to building and maintaining a full-blown storage system. Leveraging cloud services to house your data off-site enhances efficiencies, increases flexibility, improves security and cuts cost through reduced personnel and hardware investments. The most common uses of the technology are cloud file storage and cloud back-up, but if you think of the cloud only in these terms you may be missing a much bigger picture.
First, let’s look at why the cloud might be a good alternative to traditional file storage systems. The pay-as-you-go nature of cloud services allow enterprises to forgo costly investment in expensive on-site storage and the inevitable replacement cost of this hardware. Add in the operational overhead of managing a local system and even the least expensive hardware solutions quickly add up to a substantial investment.
In the cloud, multi-tenant infrastructure operates at massive scale and since you only pay for what you use your cost savings can be significant. But beyond pure costs, storing data in the cloud is worth considering because it gives you greater flexibility, improved access to your data and most likely improved protection of data well beyond what might be able to do in an in-house data center. These benefits make cloud storage very attractive, particularly for data life cycle tasks like deep archive, disaster recovery and backup and data storage behind web applications. At the same time, lots of vendors are vying for a piece of the pie and trying to make cloud storage look and act like traditional on-site storage (i.e., NAS or iSCSI) in the data center. These gateway technologies simplifying moving your data to the cloud and let you manage it the same way you’re using data center storage today.
Of course, there will always be the enterprise naysayers that claim their data is special and requires extreme levels of protection that can only be assured by retaining the data in a local data center. The reality is that most cloud solutions provide far greater data protection than you could afford on your own. Most commercial cloud service providers today provide multi-site replication of data often with at least three copies. This replication is built on an infrastructure that is built to ensure data integrity at an extreme scale, far different from what your enterprise is likely to encounter in its data center. Dealing with data integrity and protection issues at such massive scale almost assures you that you’ll find your best data integrity in the cloud.
Still, there’s a lot more to cloud-based data storage than simply replacing traditional storage systems. Because of the way cloud storage is accessed–over the network via web services API’s–it has become programmable storage. Or at least the access to data stored in the cloud is programmable and that is a very big deal. Data in the cloud is typically enhanced with meta-data–data about the data–that makes it easy to tag, share, search for, use in collaboration, present in multiple views and make it accessible to the latest generation of applications designed to work with data from multiple sources. To many enterprises this is a completely new way of thinking about their data.
The underlying protocols that enable this new data model are relatively simple. REST (Representational State Transfer) is a web-friendly architecture utilizing self-contained web services to be used by other applications over the network. REST is a stateless architecture, so RESTful services handle each request as independent transactions, unrelated to any prior requests. Virtually all the new web-based applications–including social networks like Facebook and Twitter–use RESTful web services to provide programmable access to massive amounts of data stored in the cloud. The result is new, innovative ways of organizing information, backing it up and delivering on-demand to any user in the network. This programmable storage model is finally enabling a true service oriented architecture where applications, data and presentation layers are truly separate and built using reusable service components.
Despite these benefits, most enterprises will be slow to move all their data to the cloud. There is still too much anxiety about giving up control of mission-critical data as well as about the reliability of cloud-based storage services. It’s one thing to back up data off-site to be “restored” in case of local disaster and a completely different thing to cede day-to-day control of the files and data needed to keep the enterprise running. Nonetheless, we’re confident that years from now, enterprise technologists will view storing data on a local storage array the way today’s teenagers view a vinyl LP as a means for storing music.
This post sponsored by the Enterprise CIO Forum and HP.
Leave a Reply
You must be logged in to post a comment.