It's time to pay attention to storage power use: Best Practices

Power and cooling isn't just a problem for the data center. According to Gartner Inc., storage managers place power consumption in a three-way tie for last place in terms of their concerns, a clear example of organizational misalignment.

This article can also be found in the Premium Editorial Download: Storage magazine: Salaries rise, but storage jobs get tougher:

Power and cooling isn't just a data center problem. Try these eight tips for reducing consumption.

as IT organizations continue to grow in size and complexity, an inevitable challenge is keeping various parts of the company from working at cross purposes. Keeping groups in synch and aligned means having common goals and metrics. This becomes an even bigger challenge when reaching beyond IT.

Such is the case with issues relating to the data center. Much has been made of data center power and cooling consumption and limitations that have only been exacerbated with the rise in energy prices. Organizations such as the Uptime Institute talk about a crisis in the data center, while research firm Gartner Inc. reports that data center managers rank power and cooling as their top priorities. However, for many organizations there's a wide gap between the chief concerns of data center managers and those most important to IT directors. Each one is guided by different metrics and, for the most part, tends to march to a different drummer.

Therefore, it's not surprising to find that while data center managers see power and cooling as a major concern, IT infrastructure managers tend to rank it low in importance. Again, according to Gartner, storage managers place power consumption in a three-way tie for last place in terms of their concerns, a clear example of organizational misalignment. This contradiction is understandable, since for years the metric by which IT has been "taxed" for data center usage is floor space, not power consumption. As a result, vendors have met their customers' demands by providing more densely packaged servers and storage that occupies less floor space. However, these products also had the unintended consequence of increased power and cooling requirements.

Organizations are becoming aware of the data center crisis and are taking steps to bring IT shops and facility infrastructures into synch. Some organizations, perhaps most notably Microsoft Corp., have undertaken initiatives to make data center cost allocation a function of power. The primary targets in these initiatives--what we call the low-hanging fruit--have been servers. However, with the increased adoption of virtualization and more efficient physical server designs, the focus will inevitably shift to other areas of the data center, including storage.

This isn't as ominous as it may sound. Projects that many shops have started or completed to better manage data on storage devices can help reduce power and cooling costs. Here's a starter list of things that can be done to help get storage energy usage under control.

@pb

  1. Avoid overspending and overprovisioning.
    It's pretty obvious that spinning disk is the source of most storage power consumption, and unused spinning disk represents wasted energy. But there are several challenges in growing storage incrementally. The first relates to the organization's ability to accurately forecast storage capacity needs. Beyond capacity planning, it requires a relationship with a vendor who can support the incremental storage growth. Finally, the technology must be such that storage can be expanded easily and with minimal disruption.

  2. Revisit your tiering strategy.
    Small, fast disks demand more energy than large, slow disks, so the distribution of data across various tiers of storage can have a big impact on your power consumption. An EMC Corp. study indicates that storing a terabyte of data on a 7,200 rpm 1TB SATA drive is 94% more efficient than storing it on a 15,000 rpm 73GB Fibre Channel drive. This provides an added incentive to ensure that storage service levels and their associated tiers are properly aligned based on application and data value. There are many situations where tiered storage allocations are far from ideally distributed. Understanding this distribution and developing a clearly defined set of service-level requirements to apply to new applications and existing apps can lead to substantial savings in equipment cost and energy use.

  3. Revisit RAID policies.
    Another facet of a tiering strategy is the RAID protection policy applied to a given tier of storage. It's not so prevalent these days, but overprovisioning of RAID 1 or RAID 10 increases the number of spindles and power consumption. When additional performance or availability isn't required, drive count can be reduced.

  4. Consolidate storage arrays.
    The more frames in a data center, the more power that needs to be reserved for them. In addition, older arrays tend to be less efficient than the latest models. Reducing the number of storage systems is an obvious option, just as consolidating physical servers through virtualization is. Speaking of virtualization, newer arrays may offer enhanced functionality such as thin provisioning to improve utilization and further reduce energy consumption.

@pb

  1. Establish power usage metrics.
    Historically, power specifications entered into a storage infrastructure discussion only during installation planning. Even within organizations with relatively sophisticated cost modeling and chargeback systems, specific metrics relating to power were nowhere to be found. However, to truly align data center and IT infrastructure objectives, this will need to occur. Data center managers focus on kilowatts and BTUs, with consideration given for related factors such as peak load times. Organizations like The Green Grid promote standardized data center efficiency metrics. It will also become necessary to establish metrics such as GB/kW or IOPS/kW, and to then determine these rates for each tier of storage as well as usage for the storage infrastructure in total.

  2. Consider solid-state drives (SSDs).
    While SSDs are still a new phenomenon for many, they can play a role in replacing power-hungry, high-speed, low-capacity disks. In addition to offering higher performance, EMC reports that on a per IOPS basis, flash disks require 98% less energy. It's therefore imperative that organizations do their homework and understand actual performance requirements before making any substantial investment in this technology.

  3. Consider massive array of idle disks (MAID).
    At the other end of the performance spectrum is MAID technology. We know a significant amount of data currently stored on spinning disk is accessed infrequently. From an energy-consumption standpoint, the most efficient disks are the ones that aren't spinning at all, and that's the rationale behind MAID. For archival data that still requires accessibility, this technology represents an attractive alternative to conventional, continuously spinning, nearline storage.

  4. Make energy usage a buying consideration.
    As the "green" demand grows, vendors have realized that energy efficiency can be a competitive differentiator. It makes sense to factor this into equipment purchasing criteria, and to consider energy impact when architecting new storage infrastructures. Often times, the focus is on capital expenditures (CAPEX) and insufficient attention is paid to operation expenditures (OPEX). However, evidence is mounting in the server world that these lifecycle costs, including power and cooling, can actually overshadow CAPEX. This isn't yet the case with storage, but OPEX, including power, is certainly a significant component of total cost of ownership. Effectively managing the power and cooling demands in the data center is a sum of many parts. Within each part, a series of improvements will combine to provide important dividends. For many, there may not yet be a sense of urgency surrounding power and cooling, particularly with regard to storage. But data centers around the country are beginning to feel the pinch. It's not simply an issue of rising costs. It's a matter of data centers and utility providers reaching production limits. When planning for future storage environments, it's an issue that will no longer be ignored.

This was first published in November 2008

Dig deeper on Enterprise storage, planning and management

Pro+

Features

Enjoy the benefits of Pro+ membership, learn more and join.

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

-ADS BY GOOGLE

SearchSolidStateStorage

SearchVirtualStorage

SearchCloudStorage

SearchDisasterRecovery

SearchDataBackup

Close