As attractive as the on-demand storage opportunity seems, launching a cloud storage service is far more complex than putting a new Web front-end on an existing service and calling it "cloud."
Although many enterprises are considering private cloud storage and other virtualized on-demand storage options, too few have the internal skill sets to manage these environments. Only 42% of IT pros said they believed their organization had adequate in-house expertise for managing virtualized storage, according to a 2011 survey of 1,000 enterprise IT managers commissioned by storage vendor EMC. A scant 18% thought they had the internal capabilities to plan and administer a cloud storage environment.
Striking the right balance between cost and capacity management can be exceptionally difficult for customers trying to keep up with all this data, which is growing at a 50% compound annual growth rate, according to some estimates. At the same time, increased regulatory compliance requirements around archiving and analytics are pushing enterprises to retain data for longer periods of time.
This ready-made prospect pool presents a prime revenue opportunity to cloud providers that have the infrastructure scale and virtualization expertise to develop agile, well-priced and stable on-demand cloud storage. These services -- which can leverage a cloud provider's existing data center infrastructure and any excess capacity -- can become a relatively easy upsell opportunity for current clients.
But as attractive as the on-demand storage opportunity seems, launching a cloud storage service is far more complex for providers than simply putting a new Web front-end on an existing service and calling it "cloud." Likewise, cloud storage management requires a new strategy for keeping cost, capacity and performance in check.
Cloud storage management requires accurate forecasting
While there may be nearly as many definitions of cloud storage as there are people to debate them, most agree that any cloud service shares some common elements defined by the National Institute of Standards and Technology: on-demand self-service, resource pooling, rapid elasticity, broad network access and usage-based billing. Combined, all of these factors add up to what can be highly adaptive and efficient storage solutions. They also highlight the challenges of cloud storage management.
As cloud storage gains appeal with cost-conscious enterprises, cloud providers are running into complications that go with the territory of structuring an efficient on-demand service that makes optimal use of capacity. These cloud storage management challenges start with the need to accurately estimate resource requirements in order to make adequate infrastructure investments that meet demand without over-provisioning. Accurate forecasting will help cloud providers not only minimize capital investments but also lower operating costs by minimizing space, power, cooling and support costs.
Providers should also have cloud storage management and automation tools in place that allow them to provision new systems quickly and efficiently.
There is good news on the cloud storage provisioning front, however. Although some providers are concerned about having to support the rapid-fire provisioning required by any cloud service, cloud storage capacity demands are likely to be more linear and predictable than those required in Infrastructure as a Service (IaaS) environments. This should make it easier for cloud providers to build the appropriate resources for their cloud storage services.
Storage virtualization key to cloud storage management
Cloud storage management is about more than making an adequate infrastructure investment. Cloud providers need to have other elements in place to support reliable and scalable on-demand storage solutions, starting with virtualization software that allows for more efficient pooling of storage resources. This storage virtualization software enables providers to consolidate storage systems, guaranteeing the right balance of efficiency and scale for cloud storage management. This not only helps cloud providers condense the number of physical systems in their environment but also lessens support and operational costs.
More on cloud storage management
Showback vs. chargeback: Showback new resource analysis tool of choice
How to make cloud storage pricing both economical and profitable
Getting into cloud storage services: Peer advice from an MSP
Cloud providers also need to have tools to optimize resource utilization. Capacity utilization methods include everything from compression to data deduplication to storage snapshot technology. Each of these cloud storage management methods has an appropriate use case.
Standard compression, which works by condensing the data size, is best used for non-critical applications because some data can be lost during the encoding process. In data deduplication -- a type of compression -- portions of stored data are compared to previous versions, and redundant bytes are removed. Unlike standard compression -- which is designed to look at small data strings -- data deduplication is targeted at larger file stores.
Both types of compression are designed for standard backup applications. For disaster recovery purposes, cloud providers may opt to use snapshots, which capture the most recent state of the data and record only the changes.
Service providers that design their cloud storage management and infrastructure to support efficient resource utilization have an operating model that can support high-growth services. By structuring their foundation in a way that anticipates ongoing resource demand, cloud providers can stay ahead of the capacity curve.
About the author: Amy Larsen DeCarlo is a principal analyst at Current Analysis, where her research focuses on assessing managed and cloud-based data center and security services.
This was first published in March 2012