Building Private Clouds

sadagopan Singam

Enterprises of all sizes and across virtually every industry are evaluating cloud computing initiatives for various reasons.


Some need to take advantage of the ability to classify the expense on the balance sheet as a capital expenditure (or one that will obtain future benefits and can be paid for by external stakeholders) versus an operating expenditure (or a current operating expense that is deducted from a company's revenues), and others are looking at streamlining support and maintenance requirements of their IT services.


From a CEO's perspective, cloud computing can reduce the time for "go to market" and reduce capital expenditures.


To a CIO, it can provide:


  1. Quick scale-up and scale-down options.
  2. Savings from maintenance and upgrades.
  3. A self-service model.
  4. Increased control.


While cloud computing has been presented by some as a panacea, there are certain aspects of this technology deployment model that are both difficult and confusing to many CIOs.


A few questions that we hear frequently from CIOs include:


  1. What will happen to my existing assets?
  2. Who ensures adherence to regulation and compliance for my IT environment when deployed in the cloud?
  3. How can I ensure security and privacy critical data in the cloud?


Public clouds can generally provide an easy, economical and scalable platform, but many organizations still have reservations with this approach due to issues with accountability, SLAs and compliance. Public cloud SLAs and contracts are often vague, data is turned over to, and controlled solely by, the provider, and it is hard to switch providers once data has been set up in a public cloud. In addition, there is always a risk that data can be lost if it is not properly backed up during the migration of the data from organization to provider.


As a result, many organizations are now seriously evaluating an alternate delivery model, private clouds, which allow for more control, and the ability to enforce accountability. Private clouds are rapidly becoming the basis for how cloud computing will be addressed by SMBs and large enterprises. Therefore, many providers are bringing private solutions to the market today.


What organizations are looking for from a private cloud implementation is participation by different internal departments that can subscribe and participate in its usage.


Today, the discrete hardware and software infrastructure, convergence, virtualization, service-oriented architectures and extensions, automated provisioning, and unified service delivery have all made cloud architectures technically and operationally feasible and very attractive to business.


This setup helps bring unified service delivery, network, computing, storage access and virtualization resources into a cohesive system and allows service providers to quickly deploy a technology foundation for creating and delivering IaaS services. This foundation not only helps bring IaaS cloud computing services to deliver results quickly, but more importantly, serves as a foundation for other services or for new services that emerge as the cloud computing market evolves.


Here are a few critical points to consider before adopting a private cloud:


  1. Maturity of Virtualization in the Enterprise: Cloud adoption requires an approach that covers the virtualized data center architecture and the cloud operations management architecture. With a virtualized data center, service providers can meet variable demand by quickly deploying servers, increasing hardware utilization rates, and enabling virtual machines to run across a physical layer of computing resources.
  2. Maturity of ITSM Processes Implemented: Cloud operations management includes technologies that run an end-to-end workflow - usage-based charge-back mechanisms, service level agreement (SLA) management and a federated configuration management database (CMDB). The data center and management architectures work in conjunction with each other. Through cloud service orchestration, changes and updates are made simultaneously to both. In essence, this results in a modular cloud infrastructure solution with data services that can be turned on and off based on customer demand and available capacity.
  3. IT Organization Culture: The adoption of a private cloud essentially changes how services are delivered internally. With provisioning and chargeback mechanisms enabled, coupled with real-time monitoring, a sea change in the nature of usage gets noticed inside mature enterprises adopting the cloud.
  4. Organizational Maturity: Is the organization mature enough to adopt the ownership model for cloud?
  5. Application(s) Readiness: Has an application assessment been done for private cloud adoption?
  6. Employee Bandwidth: This is often a neglected part of the equation and should be carefully considered when estimating costs of the project. Corporations need to ask themselves, "Who will be responsible for monitoring and managing the IT environment in the cloud?" In a public cloud, it is the operating system that needs to be taken care of, and in private clouds it is hardware. Depending on which delivery model is chosen, there will be a difference in the way monitoring and managing works for the organizations' IT landscape.


The ideal solution to these challenges is to adopt an "On-Demand" model for cloud monitoring and management, which gives the users the option to choose services from a service catalog or to initiate cloud services on a pay-per-use (PPU) plan.


Usually there are two approaches to building a private cloud:


  1. The "Top Down" approach, which starts from service management and governance and moves down towards virtualization.
  2. The "Bottom Up" approach, which begins from virtualization and moves up to governance and service management.


No matter what approach companies take, the most important question is the aspect of service governance and IT service management in the cloud. Is ITIL still relevant in this new model? If so, what shape and form must it take? Broadly speaking, ITIL processes (like incident management, change management, problem management, release and deployment etc. ...) work the same way in a cloud environment as they work in a non-cloud environment. The processes affected by a cloud environment are financial management, service catalog management, service level management, event management and configuration management.


However, decision makers must remember that there is always an alternative to private cloud creation. One is the use of reference architecture. Reference architecture (think "cloud builder's guide") is a process in which corporations look at their current architecture and what its future needs will be by taking into account specific technologies, patterns, business models and market segments.


Whether a corporation chooses to partner with a provider to create a private cloud, public cloud or reference architecture, it should closely monitor and align SLAs and key performance indicators (KPIs) to business goals to ensure the investment will yield future financial benefits for the business.

Add Comment      Leave a comment on this blog post

Nov 14, 2011 7:48 AM Aditya Aditya  says:

Good insight into benefits of cloud computing, adoption strategies and organizational considerations prior to adoption of the cloud platform, just viewed an excellent video presentation Adoption Roadmap of cloud computing focusing on cloud adoption strategies @  


Post a comment





(Maximum characters: 1200). You have 1200 characters left.



Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Resource centers

Business Intelligence

Business performance information for strategic and operational decision-making


SOA uses interoperable services grouped around business processes to ease data integration

Data Warehousing

Data warehousing helps companies make sense of their operational data