Intel, iTRACS Redefine How IT Efficiency is Measured

Tuesday, February 7th, 2012

As we previously discussed on January 30, iTRACS made an important announcement recently. It introduced a new set of efficiency metrics called DCIM Business Output™ that is poised to take the industry closer to the holy grail of efficiency.

What is the holy grail?

The ability to measure the efficiency with which IT delivers quantifiable and continuous value to the business.

What would every data center executive love to have under his/her belt?

Irrefutable evidence of how efficiently their IT physical infrastructure is serving the goals and mission of the business.

DCIM Business Output™ attempts to measure the ratio between business output delivered by IT assets and the energy/cost required to deliver it: business value achieved vs. the energy/cost to achieve it.

DCIM Business Output™ accomplishes this by measuring business output and energy consumption at the individual device level. The iTRACS partnership with Intel plays an important role here. By integrating Intel Data Center Manager with iTRACS Converged Physical Infrastructure Management™ (CPIM™), the two leaders are broadening and deepening the iTRACS PowerEye™ strategy through the collection and analysis of power, temperature, and environmental information at the device level (CPU).

This takes IT executives one step closer to being able to compare two key metrics at the asset level:

  • Business output – what is each data center asset accomplishing for the business? (CPU cycles etc.)
  • Energy consumption – how much energy is that data center asset consuming to generate that output?

The first DCIM Business Output™ metric is Asset Efficiency, which we talked about on January 30. This compares CPU utilization vs. power usage at the individual device level, creating an opportunity to conduct asset-to-asset comparisons to determine which asset is delivering business value at the highest efficiency.

Here is the second metric, IT Efficiency, or “the PUE of IT.” As iTRACS’ Dan Fry writes in his Green Data Center: Myth vs. Reality article in Computer Technology Review, the label PUE of IT is meant to build on the industry’s growing understanding and usage of PUE.

“Just as PUE measures the ratio of facilities to IT,” Fry writes, “this proposed new metric would help determine how much energy is going into IT assets that work directly for the business (servers, etc.) vs. IT assets that support this work but do not contribute directly to it (storage, switches, etc.).”

The goal is to minimize the amount of energy consumed by assets that aren’t working for the business so you can put more energy into assets that ARE working for the business.

So what’s been the response so far to the concept of DCIM Business Output™?

Fry says it’s been very positive. “The customers and analysts I’ve shared this with like it. They are intrigued. They especially like how flexible the approach is,” he writes. “The ‘business output’ used in the metrics could be whatever each data center owner or operator wants it to be, depending on the organization’s business model and mission. An e-commerce company might measure business output as transactions per second or customer revenue per minute. A social media company may look at ad revenue or the number of social transactions. A cloud provider might look at services delivered or compute power charged to customers.”

As Fry points out, DCIM Business Output™ is not a competitive benchmark like PUE where specific numbers are targeted. It’s a comparative measurement where your company evaluates its own step-by-step improvements in energy efficiency. And gains another perspective into the business value of your data center investment.

Fry concludes, “DCIM Business Output™ is by no means offered as a definitive metric. It is a step in a new direction.  The journey is an iterative one.”

iTRACS Collaborates with Intel on New Direction in Efficiency

Monday, January 30th, 2012

What’s the holy grail of IT efficiency?  For many data center executives, it’s the ability to measure and optimize the efficiency with which the data center delivers quantifiable value to the business.

A lot of very smart people are working on this. As Wired.com recently wrote, “What’s the best way to measure the efficiency of a data center? No one really knows. But there’s no shortage of IT minds trying to figure it out.”

Recently, iTRACS, the global leader in enterprise-class DCIM solutions, announced a new set of efficiency metrics called DCIM Business Output™ that is poised to take the industry closer to a new way of measuring efficiency.

DCIM Business Output™ attempts to measure the ratio between business output delivered by IT assets and the energy/cost required to deliver it: business value achieved vs. the energy/cost to achieve it.

DCIM Business Output™ accomplishes this by using device-level data to measure business output and energy consumption at the individual asset level. This is where the iTRACS partnership with Intel plays such an important role. Under the partnership agreement, the Intel Data Center Manager software suite will be integrated with iTRACS Converged Physical Infrastructure Management™ (CPIM™), broadening the iTRACS PowerEye™ strategy through the collection, management, and analysis of power, temperature, and environmental information at the device level (CPU).

This takes the industry is one step closer to attaining the ultimate measure of efficiency by comparing two key metrics at the asset level:

  • Business output – what is each asset accomplishing for the business? (CPU cycles etc.)
  • Energy consumption – how much energy is that asset consuming to generate that output?

The first proposed metric in the DCIM Business Output™ series is Asset Efficiency. This compares CPU utilization vs. power usage at the individual device level, creating an opportunity to conduct asset-to-asset comparisons to determine which asset is delivering business value at the highest efficiency.

 

 

This “asset vs. asset” example demonstrates how a data center owner or operator might use the Asset Efficiency metric. In this example, Server B, which has the potential to deliver higher business output, is actually running less efficiently than Server A. This is a problem for both the business and the planet. Using this information, the data center might take steps to increase server utilization, or, if necessary, swap out older energy-guzzling assets for newer, more energy-efficient ones.

Next up: The second metric, IT Efficiency

iTRACS and Intel – Redefining Energy Efficiency in the Data Center

Tuesday, November 29th, 2011

Last week, iTRACS and Intel® announced a partnership agreement to enable a new level of energy efficiency in the data center. The two companies have joined forces to extend the visibility, reach, and analytics capabilities of iTRACS PowerEye™, a best-practices power management strategy that offers end-to-end visualization and management of the entire power chain inclusive of all interdependencies, across IT, Facilities, and Building Management Systems.

Intel’s Data Center Manager software suite will be integrated with iTRACS’ Converged Physical Infrastructure Management™ (CPIM™), broadening the iTRACS PowerEye™ strategy through the collection, management, and analysis of power, temperature, and environmental information at the device level (CPU).

This combined capability will enable rich knowledge-based decisions as joint customers manage real-time events, assess utilization, plan capacity, commission equipment, perform technology upgrades, and conduct other vital management tasks across the whole of the enterprise infrastructure.

“Energy efficiency is an established corporate mandate, yet management strategies have lacked a meaningful efficiency metric,” says Elizabeth Given, President and CEO of iTRACS. “We’re excited to work with Intel, a true innovator, to aggressively address this challenge. We’re combining iTRACS’ interactive 3D visualization, single-pane view of the IT ecosystem, Future View, and ‘What If?’ analysis capabilities with Intel’s deep expertise and unparalleled granularity in CPU data collection and aggregation, monitoring, trending, and analysis. It’s a compelling alliance for customers seeking a strategy to quantitatively visualize, manage, and reduce energy usage across their enterprise infrastructure.”

It will be noteworthy to watch this partnership evolve. One company is a leader in enterprise-class Data Center Infrastructure Management. The other has unparalleled expertise in the infrastructure at the core of the data center – the CPU. Makes for an interesting mix …