Despite the confusion surrounding what cloud storage really means and how it might be used, few advances in storage have been hyped as much as "cloud storage." Understandably, IT professionals are intrigued by the prospect of reducing their reliance on internal capabilities, eliminating capital spending, and improving the overall cost structure of specific parts of their storage environment by moving workloads to the cloud. At the same time, every vendor in the space is painting its products and message with a cloud veneer. To cut through the hype, Forrester defines cloud storage as:
Storage capacity that is disaggregated from the primary computing environment, with the location, ownership, and operation of the storage resources managed by a single or a combination of service providers.
While it may be tempting to think (and market) that the cloud will eliminate data center storage as we know it in one fell swoop, this isn't likely. There's no magic in the cloud, and the concept of a consistent architecture with geographic separation from the primary data center is not well suited to all workloads. Instead, buyers and vendors need to think carefully about the specific workloads they are currently provisioning for that might be effectively solved in a cloud architecture.
Slideshow: Cloud Storage Lives Up to the Hype
The majority of today's cloud storage options focus on low price, but lack the security, scalability, and trust that enterprises require. Recent research from Forrester found that currently, only three cloud storage use cases are ready for prime time:
1. Whole in-cloud applications with their own storage
Such a delivery model allows the storage and server resources to remain collocated, just not inside the walls of the firm. With this shift, the data integrity, resiliency, and backup of the application become the responsibility of the software-as-a-service (SaaS) provider, which presumably has more resources and better expertise than its customers.
2. Backup to the cloud
Cloud offerings differ from traditional outsourced backup in that they typically leverage a globally dispersed, shared-tenancy grid system that houses many customers' data together in a cost- and security-optimized infrastructure.
3. File storage in the cloud
Many firms with widely dispersed regional sites already go across the WAN to reach centralized file services, so moving this content to a cloud-based offering is not a huge stretch. It's an especially good fit for data with low criticality and low access frequency, such as archive files and older project data.
As you explore these use cases, you'll need to select vendors that provide the proper underlying facilities, infrastructure, and storage management applications. Done right, cloud storage has tremendous potential to free up IT professionals from complex and onerous storage management. Just be sure to pick the workloads that make sense and move forward carefully, evaluating the risks and rewards of doing so.
Every infrastructure vendor is desperate to capture cloud customers, but the innovative ones will be those that understand your business and provide a viable offering that solves your problems. Don't do what's going to help your vendor validate a vague new business model, do what's going to better solve the significant challenges of managing your firm's data. To get the benefits without jeopardizing operational effectiveness and introducing greater risk, you should:
Be Specific About the Workload and Use Case for the Cloud
There are limitations on what type of data will work well in a cloud deployment. You won't succeed today if the data you are working with doesn't fit one of the three cloud categories proposed above. Future improvements in computing, remote caching, or WAN optimization may allow database applications to use storage in a remote location, but this isn't a viable option in the near term.
Consider the Risk Profile of Data that Would Live Outside Your Walls
Cloud offerings are still young, so it makes sense to start out with lower-criticality data. Firms with varied data types should create a risk classification and a policy that defines which data is acceptable to store in the cloud. Offerings vary in terms of security credibility, so proceed carefully. Demand a high level of transparency from vendors as to their capabilities and exposures, in an effort to build critical trust.
Pay Close Attention to the Contracts that Govern Cloud Offerings
Make sure that vendors provide clear SLAs that describe what will be provided, the performance and availability levels to be guaranteed, the provisions in place to secure access to data, and the penalties for violation these SLAs. Good multitenancy and service delivery capabilities are not enough. Push for clear contractual obligations that govern the relationship for these critical business transactions and help set proper expectations.
Create a Contingency Plan
Your data in the cloud is only as available as the cloud service itself -- something many organizations learned the hard way in the fall of 2008 when Amazon S3 suffered a major outage. Despite any cloud vendor's best efforts, outages are inevitable, and it behooves IT professionals to have a backup plan. While this doesn't entail a full resiliency plan for your data that lives in the cloud, you should have a means of recovering and serving that data, even if in a degraded mode. After all, you have an SLA with your internal or external constituents, who won't accept you passing the buck. Consider the long-term costs of the cloud.
Don't settle for cloud vendors that essentially are providing leasing options on the same capabilities you could have built yourself. This will result in higher costs over time. Likewise, it might be attractive to eliminate internal staff and high capital costs, but chances are a cloud offering will just augment your capabilities, not replace them. Take a big-picture view to determine whether the services are more effective than those you can provide internally -- and if the total costs for the solution will be lower than the cost of managing storage internally.
Andrew Reichman is a Senior Analyst at Forrester Research, where he serves Infrastructure and Operations professionals.