Private cloud computing models alleviate some cloud security issues

Private cloud computing may provide enterprise organizations seeking the performance, scalability and cost efficiencies of a cloud architecture but have serious security concerns.

Private cloud computing seems to be surrounded by much fog and misinformation as to exactly what it means for the enterprise. What types of companies are a good fit? Who might need a private cloud? Are security issues the only reason? Developing your private cloud takes considerable skill but is a viable option for companies looking to achieve economy of scale behind the corporate firewall. So say respondents of the 2011-2012 TSS Java Trends survey.

What exactly is a private cloud? It is simply a marketing term for a proprietary computing architecture that provides hosted services to a limited number of people behind a firewall. A private cloud is typically implemented by first virtualizing internal commodity hardware, then adding an application services layer, and finally managing the new infrastructure as a single resource. This single resource is often allocated to applications that process and store sensitive and private company information in accordance with internal or government policies, such as Sarbanes-Oxley and HIPPA.

TSS readers are just stepping onto the private cloud bandwagon with 27% of survey respondents saying their organization is currently developing a private cloud computing model. The trend is increasing too, with an additional 24% indicating they are planning a private cloud over the next 18 months. Interestingly, 31% said they have no plans for a private cloud. This likely means that the critical mass required to realize a true economy of scale is out of reach for some.

Privacy and security issues still seem to be at the forefront of the decision to create a private cloud. 70% of respondents agree that cloud computing security issues are in the top three greatest challenges their organization faces when deploying applications to the cloud. In hiding complexity from customers, public cloud computing models keep the where and how of data storage and processing from the purchasing organization.

Of course, this is unacceptable in many IT organizations. Some reluctance may be generated by simple fear of the unknown, a lack of familiarity by IT management, or just a mistrust of outside IT organizations. However, compliance regulations also figure into the equation. Still, there is some evidence to suggest that cloud provider security may beat on-premise in some situations.

In addition, cloud computing, whether public or private, represents a big change in how IT is managed. Organizationally, many managers like having an in-house data center, both for practical and political reasons.

Private cloud adopters, however, seem to go hand in hand with organizations using application virtualization technologies. Fully 60% of TSS readers surveyed said they were using application virtualization. It appears to have become a standard strategy among enterprise organizations. Virtualizing applications allows them to run in an insulated server environment where each app has its own resources allocated and management of resources is easily administrated.

One excellent alternative to the popular vendors in implementing a private cloud is an Apache project called Hadoop. Hadoop is a Java based, open source programming framework for building applications that work with numerous nodes and tons of data. Hadoop provides network node location awareness and this allows Hadoop apps to process locally on the node where the data is stored. Keeping the data and the nodes doing the processing in close proximity can result in massive improvements in scalability and performance for the private cloud. Only 8% of respondents are currently using Hadoop in their organization, but they find themselves in good company as both Yahoo and Facebook use Hadoop extensively.

Private cloud computing offers both the security of traditional on-premise infrastructure and the scalability of newer cloud computing models. A private cloud may provide more reliable uptime and better tracking than public clouds because you have complete control over your network. At the same time, it promises better scalability than on-premise architecture because you can scale as the need arises. If you need the ultimate in data privacy and service integrity, the corporate private cloud approach is your best shot. There are challenges in going this route, but you have complete control on how it will all turn out.

Dig Deeper on Java cloud platforms

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.