Five steps to lower storage costs

Asian businesses can reduce their storage infrastructure costs by basing their buying decisions on the way data is used, says senior Sun Microsystems executive.
Written by Lynn Tan @ Redhat, Contributor
Organizations in Asia should build their storage infrastructure around the way data is organized and used, says a senior Sun Microsystems executive.

Joe Heel, senior vice president of Sun's Global Storage Practice, told ZDNet Asia in a phone interview that storage infrastructure buying decisions should be based on the way businesses organize and use their data.

"When storage first became a part of the computing paradigm, people paid little attention to this and bought storage in whatever was available, and usually to increase disk space storage," Heel said. "And as a result, the storage was relatively poorly utilized and the cost was relatively high."

According to Heel, storing all types of information in the same way makes little business sense, because it increases costs for a company. The amount of data that needs to be stored has also become "a very substantial cost item for many corporations", he added.

To help businesses address these storage challenges, Sun has designed the Information Management Maturity Model (IM3) which comprises five steps to greater financial accountability.

According to Heel, the model helps to address the three things that CIOs are ultimately concerned with: increasing revenue, reducing costs and mitigating risks.

"The IM3 model helps you understand where which type of data you have [is stored] and helps you create an architecture that best matches the way you want to use your data," Heel explained, adding that the model helps Sun customers save money by appropriately configuring their infrastructure to match their data needs.

The five steps of IM3 are:

1. Define business requirements
Identify unique business requirements and how they relate to data. These relate to business recovery, performance levels, compliance issues, customer expectations, and competitive needs.

2. Assess present situation
Define service level agreements (SLAs) for the identified business requirements. This makes it possible to remove or redeploy storage assets that are not needed.

3. Tier, consolidate, and simplify
A tiered storage strategy ensures that expensive primary disk is used only for high-value data. It also relieves the backup burden by relegating low-value data to less expensive, offline storage tiers.

Citing archival data, such as tax or medical records, as an example, Heel said that this type of information is best stored using devices and software that are inexpensive, but do not provide "as fast access to that data" compared to primary storage.

"For example, tape storage is a very good place to put such data," he said, noting that it is very cheap and consumes no energy when unused, but requires longer access to the data stored.

"If you know this, and if you know which data is of that nature, you can put [the data] on the right type of device and you can see how that might save you a lot of money," Heel said.

4, Define a 'catalog' of service levels
Define a catalog of service levels, each with a predefined cost.

If a business unit needs 99.999 percent availability for a mission-critical data set, it can receive--and pay for--that level of service. When the value of the data drops and 90 percent availability level will suffice, the business unit can switch to a lower service level--and pay a lower cost.

5. Monitor, manage and provision
Built-in monitoring and reporting capabilities help provide better visibility into the true costs of storage, optimized storage environments.

Monitoring and reporting are also key to implementing a chargeback model, which allows the IT department charge specific business units for services rendered, Heel noted.

Editorial standards