Cloud computing shows no signs of slowing down as a top technology trend. Since it first burst on the scene, cloud computing has been heralded for its ability to provide inexpensive computing through the use of shared resources. But as concerns continue to be raised about security, visibility, and availability in public settings, the focus of cloud computing has shifted. Increasingly, organizations are looking to deploy cloud computing internally.
According to a recent survey of the Independent Oracle Users Group, “adoption of private cloud services for IT workload processing or infrastructure” is now outpacing the use of public platform service providers.¹ The survey notes that users of private clouds “are reaping cost savings by eliminating duplication, standardizing for operational efficiency, and through higher asset utilization.”
Nevertheless, some organizations continue to hold off deploying private clouds, saying they lack the IT budget to make such a major infrastructure change.² But that doesn’t have to be the end of the story. Getting to the cloud doesn’t mean you have to “rip and replace” the infrastructure investments you’ve already made. Consider, for example, how storage accounts for a major portion of the IT budget. Are you doing everything you can to leverage your existing storage infrastructure?
Continue reading to learn what questions you need to ask about your storage consumption to take advantage of a private cloud with the infrastructure you already have.
According to the Enterprise Strategy Group, as much as 70% of physical storage capacity is wasted through mismanagement.³ This occurs partly as a result of disconnected storage management tools and processes. Rather than fixing the issue by using a product that is well integrated with internal IT processes for managing capacity, all too often administrators simply buy more and more storage to meet never-ending demands for storage capacity. Needless to say, that state of affairs can’t continue forever.
For any cloud solution to be viable, the storage infrastructure needs to be capable of delivering on the Storage as a Service model. That means the following requirements must be met:
- Are you employing on-demand storage? Cloud computing is based on the fundamental premise that operational efficiency is realized through the pooling and sharing of resources. That means using only what is needed and returning resources when they are no longer required to the pool. Today new techniques are being adopted that provide this ability to deliver on-demand storage. For example, thin provisioning challenges the long-standing storage approach of having to dedicate capacity up front, based on allocation. This results in higher capacity utilization, eliminates the guesswork in new application provisioning, and reduces capital expenditures and operating costs. Thin reclamation, meanwhile, provides the ability for data centers to reclaim storage once the space is no longer used by any application.
- Is your storage SLA-based? Just as availability is key to the success of cloud initiatives on the server side, so too will SLAs play an essential role in provisioning storage. However, providing storage on demand isn’t enough. The right kind of storage based on the needs of services and applications is also necessary. So, for example, a business-critical application might need extremely fast solid-state storage, while other non-critical applications could make do with slower storage.
- Do you tier your storage? A common practice in many legacy data centers was to allocate top-tier storage for all applications, regardless of the requirements of the application. This was often done to ensure that minimum SLAs and performance were met for any application. Such an approach no longer makes sense in a private cloud. Storage vendors today routinely include multiple storage tiers within a single array. One way to get more out of existing storage assets is to automate a tiered storage model so that less important data can be migrated out of expensive Tier 1 storage resources. Organizations can classify their data and determine its value to the business and then map that value to the appropriate physical tier. This enables organizations to ensure that they have the right data, on the right tier, at the right price.
- Do you know what storage you are consuming and the consumption trends, over time, for your entire storage environment? True cloud computing means centralized storage management and operations. That means visibility and control over a multi-vendor storage environment. Most IT organizations know how much storage has been allocated, but they lack the visibility into actual storage consumption. They don’t know what waste is out there and have no means for reclaiming wasted storage. It’s important to seek out solutions that can work with all of the storage you have now and will have in the future. Also make sure those tools have deep visibility into storage consumption as well as remediation capabilities built-in.
- Is chargeback enabled? Another fundamental premise of cloud computing is that services and resources are tracked and charged back to the consumers based on their usage. Today that’s a greater challenge than ever before. After all, pinpointing ownership, understanding data usage, and managing data consumption become extremely difficult when the amount of incoming data overwhelms the ability to maintain it. Fortunately there are now management tools that provide storage chargeback capability across heterogeneous environments, giving visibility into storage usage and enabling chargebacks.
- Are storage efficiency and reduction technologies in place? Data deduplication, compression, and auto-tiering need to be primary storage considerations. You also want to be sure you’re controlling storage sprawl by creating incentives on the demand side to reduce storage usage. In addition, streamline backup and recovery times by moving older, infrequently accessed data from production sources into a centralized archive.
Before you’re ready to implement a private cloud, Symantec recommends you make sure you’re leveraging the existing infrastructure to achieve the same efficiencies and then modernizing it as needed. That means converting static servers, storage, and networking into a virtualized pool of resources. It means replacing static provisioning with self-service provisioning. And it means implementing metering to demonstrate value to the business. In the end, getting to the cloud doesn’t mean you have to rip and replace.
- ¹ “Privatizing the Cloud: Study on Enterprise Cloud Adoption,” Independent Oracle Users Group, November 2010
- ² “2011 Private Cloud QuickPoll Survey,” IDC, March 2011
- ³ “Symantec’s Broad Approach to Thin Provisioning,” Enterprise Strategy Group, June 2010