Year of the Container Data Center: Spring has sprung for the new limb on the Data Center family tree

Momentum is a powerful force, and momentum coupled with synergies cannot be denied.

Today IBM is announcing their entry into the Container Data Center space. Starting with their newly released iDataPlex high density server product, and riding the wake of Microsoft’s widely publicized embracing of Data Center Containers, the recipe was there for a product combining the two concepts.

iDataPlex is a water cooled system of very high density computing which is said to squeeze twice as many servers as conventional systems into the same amount of floor space while reducing power demand by as much as 40 percent. This is a solution in demand by a wide range of business applications, and tracks the trend of massive utility computing platforms in the implementation phases across the globe.

Now we also have the decision by IBM to configure 40 foot containers with iDataPlex infrastructure. The trend is now undeniable, with product releases over the past months

from Verari, Rackable, Sun, et. al.,… and the high profile deployments using these solutions.

I think that the Container Data Center concept holds a lot of potential for a number of reasons. This potential is reflected in the needs I’ve seen in the Clients we serve in our consulting firm.

Modularity is one dimension of this potential. Our Clients are often planning the construction of a data center facility to accommodate long term business growth. Data centers are expensive enough, let alone building excess capacity that is not participating in revenue generation. The application of Containers facilitates an option to grow incrementally, and could reduce a large amount of planning risk.

Real Estate is another dimension. The concept of Containers gives an out-of-the-box option to plan for land without the traditional data center building. This could have a significant favorable impact on the construction budget in some cases.

Another dimension is Disaster Recovery. The Container option may be an attractive fit for many enterprise DR scenarios. For many enterprises, DR is not an inherent service that exists from the moment the Business application lights up for users. Sadly, it is often a separately funded risk mitigation project well after the fact. A Container based solution may offer options that are favorable in terms of implementation time as well as for operational costs.

A few years ago, I was tasked with addressing Business Continuity and Disaster Recovery options for Home Depot stores that were at risk for weather-related outages (ie, hurricanes in Florida and the US Gulf Coast). At that time, I collaborated with Intel, HP, Wolf Coach, and Cisco to create a straw-man design for a relocatable data center. That data center was, for that application, quite small in terms of quantity of equipment and compute capacity. I’m certainly not claiming to have birthed the Container Data Center concept, mind you, but rather adding the perspective of mobility and temporal deployments of “contained” data processing footprints.

The Container implementations in the news are largely for high profile data center projects, that would get press coverage whether using Containers or not. If you have an experience with Container Data Centers I’d greatly welcome the feedback.

New Data Center Cost Estimate Updates from the Uptime Institute

You may be familiar with the “Dollars per KW plus Dollars per Square Foot of Computer Floor” white paper from the Uptime Institute for data center construction cost estimation. Following the update to their widely referenced Tier classifications paper, this benchmark for thumbnail cost estimation has been updated as well. If you’d like to look at it directly, it can be found here.

Beginning several years ago, the Institute put together some very good work to give us guidance on the cost of data centers according to Tier classification. The cost model is based upon the premise (quite accurately so) that the cost of the MEP infrastructure is far and away the dominant driver in the cost of erecting an IT facility. The model simply separates the cost of the building and support space from the cost of the MEP infrastructure.

The benefit offered by the Institute’s work in this regard is the benchmark cost data for applying this model to your own project. The reference benchmarks are based upon a set of real-world contemporary data center construction projects, where the relevant component data is extracted, averaged, and provided for our use.

It’s the term, contemporary, in the earlier statement that makes this update of the white paper important. First, the data is updated using four additional new data center projects,… so the sample set is that much larger. This might seem insignificant, but besides having four more data points we also have four new recent data points.

Over the course of the past 12-18 months, there has been a sharp escalation in the cost of materials and components critical to the construction of data centers. The cost of steel and copper has increased world-wide, by over 50% during this time. There has also been a boom in construction of very large data centers globally, which is driving the cost of critical components like generators and chillers. Recent developments such as these are folded into the guidance given by the Institute’s model.

Not stopping with the data updates, the Institute has taken the opportunity to also include cost estimation data for “empty space” in their model. This is a pragmatic addition, given that a facility owner will want to include space in the new building that might not be specifically programmed, but is likely to be used for growth or operation of the raised floor.

Finally, the Institute offers another piece of pragmatic guidance, suggesting the level of “seed funding” necessary to accomplish the planning needed to arrive at an accurate budget estimate. This seed money is used for project planning, design/build contractors, and consulting necessary to bring the definition of the project into focus sufficient to confidently go to the Board for approval.

How Cold Should My Cold Aisles Be?

I was recently asked a question about how cool the cold aisle should be in a data center using hot/cold aisles to manage temperature.

This question is difficult to answer without other background information.  I’ve seen specifications reference ASHRAE guidelines, Uptime Institute recommendations, and so on.  For me, TIA-942 (which is the basis for the international standard on data center design), is a good framework to lean upon.

TIA-942, Appendix G states the following for the capabilities of the raised floor area environmental air:

• 20 degrees C to 25 degrees C (68F to 77F)
  • Normal set point: 22 degrees C (72F)  +/- 1 degree C (2F)
• Relative Humidity: 40% to 55%
  • Normal set point: 45% RH +/- 5%

Now, all that may be easier said than done, of course.  The whole point of the cooling is to keep the equipment happy (operating in the prescribed temperature range).  Heat is generated by the equipment’s consumption of electricity, the level of utilization, and the density of the technology used in your data center (which may also vary by location on the raised floor, creating a non-uniform temperature pattern). 

This is also modulated by the power density capabilities of the data center’s MEP infrastructure.  If you have a relatively old facility, chances are that the power density capabilities of the facility are less than 50 Watts per Square Foot (W/SF).  If you have a modern facility, you may enjoy capacities of 85 W/SF, 100 W/SF, or even 150 W/SF (higher than that are somewhat rare these days).  A well designed MEP infrastructure has cooling capacity matched to power density capacity. 

Regardless of that though, it’s possible that you may use technologies that squeeze a very high amount of power consumption (and heat generation) into a small area.  Hot/Cold aisle arrangements may not be enough to cool those spots, and may cause consideration of supplemental cooling on the raised floor.  There are several in-row cooling technologies with which you can place a floor-standing cooling unit in-row with your equipment racks to deliver extra cooling capacity to selected areas.  One should be careful when using supplemental approaches though, that the redundancy of MEP infrastructure supporting your intended Tier Level is not obviated by such an implementation.

Do you have any experiences with specifying hot/cold aisle metrics that you’d like to share?

 

 

My Kingdom for a Nice (big) Raised Floor!

‘Ever notice how difficult it is to find good data center space today?  It wasn’t all that long ago that I recall walking through (seemingly) acres of raised floor in mostly vacant hosting centers, being courted by the local sales team to place my infrastructure in their space. 

What a difference a few years makes.  Today, it’s a Seller’s market when it comes to raised floor space.  The market is especially tight if you’re looking for larger amounts of contiguous space.  If you’re looking to plant, say 10,000 or 20,000 square feet of racks at a collocation provider today,…. Fa-get-a-bah-dit (well, almost).

 

 

 

 

 

 

There’s an additional complication though.  Besides looking for a quantity of IT space, enterprises today are also interested in Tier-3 or higher space.  Hosting customers want the extra “9” in the availability number, and so to their Customers.  If a higher Tier rating is required in your search, the pool of candidates is likely limited to those facilities constructed over the past few years.

 

 

 

Let’s not forget too, the cost component that I sometimes call “The Infrastructure Tax.”  Hosting providers have had to find a way to pass on the (steeply) rising cost of power and cooling to their customers.  In Atlanta where my office is located, power is still relatively cheap, but in other parts of the world where we serve our Clients, the cost of power and the margins applied by the hosting provider can be quite hefty.  In some cases, the all-in expense of using a collocation provider can compete with a build/own approach from a 5-year ROI perspective.

 

 

If you’d care to share your recent experience with finding high-Tier collocation space in your part of the world I’d be delighted to hear about it.

 

I love the work I do for my Customers

I am passionate about the work that I do for my Customers. I work as an IT/Business consultant, for medium to large companies across a variety of vertical markets. Though each of my Customers has unique businesses and unique concerns, in all cases there is a need to progressively and opportunistically close the alignment gap between the business and IT.

We live at a time in which the underpinnings of IT services are, or are on the path to be, commodities. Thank goodness for open standards and mature governance frameworks. The Business has options now. IT organizations that have embraced these opportunities can count themselves as accelerators and enablers for the Business they serve.

<!–[if gte vml 1]&gt; &lt;![endif]–><!–[if !vml]–><!–[endif]–>

Across the board, my Customers are very smart. They understand the business that they serve. They understand the new technologies, and are beyond the “hype” phase of the technology adoption curves. They recognize and face the challenges before them and they show a strong sense of urgency in their desire to improve the levels of IT services in their firm.

As a consultant, I serve my Customers in the areas of IT Governance, IT-Business Alignment, IT Strategy and Planning, Data Center consolidations, BC/DR, IS Security, and strategic roadmaps for specific infrastructure technologies.

This is truly exciting work. This work has a direct and positive impact on the value of IT to the Business. I can’t think of a time in my career when the work I do is as impactful to the Business and its Customers. These are exciting times.