ESG Lab tested the iSR 6200 series and documents the results in a just-published  Lab Validation report that found it to be a flexible, easy-to-use, and high-performance universal data migration tool.

See also:  ESG Brief: QLogic iSR Accelerates Data Migrations

See also:  ESG Lab Report: QLogic iSR 6200 Series

While this is all pretty cool, the most interesting use case that I heard came from an end-user and had NOTHING to do with Data Migration (other than avoiding it altogether).

This end-user has month-end batch processes that require huge CPU, memory, and storage performance for just a few days each month. For the rest of the time, they need to have access to the data, but nothing much else happens. For less than one week a month, these apps live on big, honking, multi-core servers with massive amounts of RAM and 8G FC connectivity to enterprise storage arrays . Once processing is done, the VMs are moved using vMotion to 1u servers with minimal configurations and iSCSI connectivity only, freeing up resources on the big machines for other critical apps. They keep an iSR 6250 permanently inline to route between the iSCSI and FC SANs, so they don’t have to move the large dataset along with the VM.

While this is not a scenario we tested in the ESG Lab Report, it uses the validated capabilities of the iSR in a clever way to provide real operational efficiency.

Solutions like this is why technology will always be cool.

Tony

How efficient, you ask?  On a typical day it has a PUE of about 1.2.  And as far as innovations go–well, there is a patent pending for the design work they did  with control of air flow by measuring differential pressure between the cold rooms and hot aisles. Again, pretty impressive for a company that makes its living building storage.

Another remarkable feature of this data center–they built it all with off the shelf components. No custom million dollar components, just stock products configured in a very efficient manner.  It just goes to show that organizations can build cost-effective and energy-efficient data centers when empowered to do so.

You can read Bob’s other blog entries at Data Center Continuum.

via Data Center Management: Going Lean and Green.

Overview

For most IT organizations, power is taken for granted—until they run out of it or have to pay for it. Organizations have typically relegated power and cooling issues to the facilities department, with IT only providing input on an as-needed basis. IT’s primary concern is focused on service levels and availability, not power consumption. However, as organizations continue to build out high density, virtualized compute infrastructures, power and cooling can be a serious issue. This is especially true as organizations continue to consolidate data centers. HP and IBM are great examples, going from 85 to six and 123 to seven, respectively. This is a trend that ESG research[1] has validated, with “consolidating data centers” ranking third on the top IT initiatives for enterprises in 2010. As the amount of power required to operate these consolidated centers continues to grow, some organizations are finding themselves in a precarious situation. In a number of geographies, organizations are being told that their energy consumption is being capped. Unfortunately, their businesses are not slowing down: the number of interactions and the amount of data continues to grow. Properly supporting the business requires more infrastructure and, hence, more power. Yet, running out of power is not an acceptable outcome. So, if an organization can’t get any more power, it needs to optimize what it does have.  In order to do that, proper tools and the cooperation of both the IT and facilities teams are needed.

Analysis

How did we get into this situation? Much like the all-too-familiar over-provisioning that occurs in IT departments when bringing a new app online (everyone adds a little headroom to the request, from Servers to Storage), the same thing has happening in facilities. In order to mitigate risk, organizations have provisioned power and cooling based on the maximum load of current or future deployments of IT infrastructure based on manufacturer specifications. However, as many of us have learned, most compute and storage infrastructures are not operating at anywhere close to maximum load and vendors tend to be overly conservative with their numbers. As a result, there is a lot of room for improvement.

Unfortunately, in order to get any measurable improvement, organizations first need to gain visibility into these environments. Most organizations don’t have the tools to deliver effective management and optimization of their power and cooling environments. While IT is very familiar with the concept of monitoring and measuring the efficiency of their IT equipment, they in fact spend millions on management software every year to gain better insight into their virtualized and physical server environments, networks, and storage—all with the goal of more effectively managing their infrastructures. The benefits derived from deploying these solutions include driving up utilization rates, accelerating provisioning, mitigating risk, and reducing costs. Facilities teams now need to invest in these tools and could benefit from the experience IT has had in purchasing, deploying, and integrating these tools. And, while the facilities team typically owns the budget for this type of expense, IT should be deeply involved as deploying a complete solution will require cooperation across the entire data center. Typically, this type of cross-domain cooperation can be facilitated through executive-level sponsorship.

Fortunately, there are a number of solutions on the market to help organizations monitor and measure the power and cooling demands of the data center. Organizations should look for solutions that:

• Monitor IT infrastructure as well as power and cooling equipment. The level of detail may vary on this, so it is important to determine what is being monitored and how. For example, is it PDUs, top of rack, or will the solution monitor actual usage at the server or even desktop level?
• Deliver compelling and actionable reports. Ideally, this would include separate reports for IT, the facilities team, and executive management. It might include items like inactive servers, storage to be targeted for decommissioning, or chillers running at 100% when it is not required.
• Auto-discover all components. While this would be great, you may find that many solutions will still require importing of initial and changed data manually. Make sure to understand the resource commitment required to maintain and update the database. In some instances, sensors may have to be placed throughout the data center or in agents loaded on devices. Again, determine the level of commitment that will be required to maintain the environment.
• Automate or offer the ability to automate based on policy. This would include the ability to turn off desktops when not in use and automatically control chillers based on need. Certainly, anything that does not require manual intervention drives greater efficiency, but users need to be comfortable with the actions taken.

There are a number of organizations providing energy management solutions including Aperture (Emerson), FieldView, Sentilla, Siemens, PowerAssure, Tivoli, Viridity, and Visual Data Center. Viridity offers a free version of its EnergyCenter that can be downloaded from its website.

By leveraging one of these solutions to monitor and optimize power and cooling, organizations should expect to see several benefits including:

• Extending the useful life of the data center. By driving greater power utilization and efficiency, organizations can breathe new life into a space previously considered “full.”
• Deferring capital costs. By making better use of an existing data center, organizations can postpone or completely eliminate the need to expand an existing site or build out a new facility.
• Reducing operational expenses. ESG research shows that for the last two years, this has been the number one justification for making new IT purchases. With the cost of power increasing in many geographies, organizations could realize substantial savings by optimizing power.
• Increasing revenues and profits. This is especially true for service providers or hosting sites. Creating more space on a valuable raised floor and increasing the amount of equipment that can be used (and sold) will increase the revenue per square foot of the data center.

What Should You Do?

A quote often attributed to Albert Einstein reads, “The definition of insanity is doing the same thing over and over again and expecting different results.” Unless organizations become more proactive in the monitoring and management of power and cooling, the data center will continue to waste power needlessly.

The first step is getting a clear picture of where power usage and cooling are today. Organizations need to benchmark the existing environment and identify the biggest areas of waste. The next step is to ensure continuous (or as near continuous as possible) monitoring of the environment. A point in time snapshot is effective to a point, but in order to optimize while minimizing risk, continuous monitoring is essential.

The initial investments do not need to large; as mentioned above, some vendors offer a free version to do a quick assessment and get a better understanding of how the organization would benefit from these solutions. Once that baseline understanding is there, organizations can budget and purchase solutions that best fit their needs.

The Bigger Truth

Now is the time to act. Organizations have spent millions consolidating data centers and deploying innovative virtualization technology to create more agile and efficient computing environments. The same can be achieved for power and cooling—extending the life of existing data centers, optimizing new ones, and mitigating risk by being able to effectively forecast future demands.

Organizations will be able to reduce operational expenses, defer capital expenses, and operate more efficiently without incurring any additional risk, plus company executives will have a great “Green Story” to tell while the business saves a ton of greenbacks. In order to accomplish this quickly and seamlessly, IT and facilities teams should be like-minded in this pursuit. It will provide a benefit to the business and will definitely reduce operating expenses. Executives with responsibility for both should be fostering this teamwork and making sure any petty issues about budgets are taken off the table.

As the amount of power required to operate these consolidated centers continues to grow, some organizations are finding themselves in a precarious situation. In a number of geographies, organizations are being told that their energy consumption is being capped. Unfortunately, their businesses are not slowing down: the number of interactions and the amount of data continues to grow and to properly support the growth,  businesses require more infrastructure and hence more power. Yet, running out of power is not an acceptable outcome. So, if an organization can’t get any more power, it needs to optimize what it does have.

In order to do that, the proper monitoring tools and the cooperation of both the IT and facilities teams are needed. To expedite the process, it would help to have the support of the executive team as well. Unfortunately, many organizations continue to spend hundreds of millions of dollars building out new data centers to optimize IT infrastructure and neglect energy optimization.

A quote often attributed to Albert Einstein reads, “The definition of insanity is doing the same thing over and over again and expecting different results.” Unless organizations become more proactive in the monitoring and management of power and cooling, the data center will continue to waste power and money needlessly. Also, given the need for highly dynamic elastic compute environments the result could be even worse–critical business applications could be at risk if a sudden rise in demand outpaces the available power.

Why wait? Now is a perfect time to act. Technology exists to create more agile and efficient computing environments. The same can be achieved for energy management, but organizations (that means Facilities and IT, with executive support) need to work together to accelerate the adoption of a solution that best fit its environments. The benefits of doing so–extending the life of existing data centers, optimizing new ones, and mitigating risk by being able to effectively forecast future demands–far outweigh the costs of implementing these solutions. Plus, companies like Viridity offer free versions of their software that can be downloaded from their website..

Other vendors with solutions include Aperture (Emerson), FieldView, Sentilla, Siemens, PowerAssure, Tivoli, Viridity, and Visual Data Center.

To read more, check out my brief on the topic.

You can read Bob’s other blog entries at Data Center Continuum.

via Green Storage – The Present and the Promising Future — EnterpriseStorageForum.com.

Appropriately called HP Flexible Data Center or Flex DC, as I have seen it referred to, it leverages a modular design of what HP refers to as a “Butterfly,” with a core building in the center and four quadrants of data center space. Scale is achieved by adding quadrants or creating multiple “Butterflies,” essentially creating a campus of data centers. HP claims significant cost savings and a turn time of less than a year. Not too bad when you consider these are real brick and mortar data centers and not containers or pods. Will be interesting to see if this is the approach HP will take with its ES (EDS) data center consolidation effort and how customers respond to this standardized approach. Would be great to hear from any customers that are going down this path.

Read Bob’s other blog entries at Data Center Continuum.