Ongoing Maintenance & Monitoring

It is an obvious truism that given enough time everything will fail. The only tool we have at our disposal to hopefully delay this eventuality is maintenance service. Unfortunately, it’s another truism that for most small computer room operations, this vital step is not performed. As is unfortunately typical, we often put off short-term inconvenience for future unplanned and unpredictable grief.

Whatever the circumstance, there are plenty of tools to simplify the organization, planning, scheduling, and performance of field preventative maintenance. Real-time monitoring systems can serve as the front line of defense against unplanned outages.

PTS emphasizes utilizing IP and Web technologies to oversee and control critical support systems from just about anywhere. For power monitoring, we prefer to take advantage of the growing trend in the power strip manufacturing industry by having clients deploy power strips that can measure at the receptacle, therefore at the device level. For example, simple alarming of support infrastructure as critical load values approach predetermined thresholds will prevent against failures due to overload conditions and therefore curtail availability stripping outages.

In any case, monitoring systems for IT attributes and physical attributes should provide proactive management and enable the quick assessment of your present situation and notify the appropriate personnel should situations that threaten availability.

Is preventative maintenance high enough on your data center to-do list? What technologies do you rely on to monitor your critical systems?

Receptacle Level Load Monitoring & Control

Power monitoring and control at the receptacle or rack level is a hot topic lately. Part of the interest can be attributed to the lure of the unknown – that feeling of “I’m not sure why I want it, but I’ll probably need it!” But there are some really solid reasons for data center managers to consider receptacle-level power monitoring/control solutions.

The ability to trace watts information at the power strip level gives a much clearer picture of how much power a data center consumes. If I have an under-performing asset, it’s easy to earmark for replacement if the problem can be measured down to the receptacle level. If an asset is under-utilized, it can be easily targeted for virtualization.

There are a number of products that can be used for receptacle-level power monitoring and management. Take, for instance, the RPC series of power management solutions from Baytech. These units let you manage power more efficiently by remotely turning on/off receptacles or rebooting unresponsive equipment. (You can read more about Baytech’s products in “Better Monitor & Control Power” at Processor.com.)

Raritan offers Remote Power Control (RPC) units that allow you to control power usage at the socket level. The units have individual LED indicators for each receptacle and, in the case of an outage, offers receptacle status retention so that power is restored only to those assets that were on previously.

There are also the Synaptix™ power distribution units from Epicenter. These products come in a variety of receptacle configurations, offer the ability to measure consumption at each individual receptacle, and can be accessed remotely.

It will be interesting to measure the true impact of these units on data center power efficiency. Don’t be surprised to find me writing a white paper on the use of receptacle-level power solutions in the coming months.

New White Paper on Power Usage

Managing data center power usage is critical due to rising energy costs and diminishing supplies. But where do you start?

To help answer this question, PTS Data Center Solutions in collaboration with Raritan, a leading manufacturer of power management products, developed a new white paper that examines the myths and realities of power usage in the data center.

Entitled “Power Moves: Understanding what you know - and don't know - about power usage in your data center”, the white paper shows IT professionals how to calculate data center power efficiency and set standards to align with the Green Grid. Its findings are based on a series of tests which examined the effects of heat, airflow and power usage in a working server environment using 3-D CFD software and intelligent power distribution units (iPDUs), among other devices.

The hypothesis was that, by knowing more about their real-time operational environment, data center managers would be empowered to manage smarter. Key findings include:

• Servers are not necessarily needed to add computational power;
• OpEx expenses can be reduced without putting required computing at risk; and
• Running servers at 80 to 100 percent can be more beneficial than running at the industry average of 60-80 percent.

To download the free white paper, please visit the PTS Media Library.

Tips for Handling Data Center Moves and Shortages of Space

Look for PTS Data Center Solutions in the July 11th issue of Processor magazine (Vol.30, Issue 28).

Kurt Marko interviewed me for the feature article, “Need More Data Center Space?: IT Managers Are Faced With Options Ranging From Simple Housekeeping To Major Construction”. Adding data center space can be a complex and costly issue. If your data center runs out of room, the basic options are 1) reorganize and consolidate to get the most out of your existing space, 2) upgrade your technology to increase density, 3) call in a contractor to renovate and expand your current facility, 4) add on a data center in a box, or 5) build a bigger-better data center. Marko’s article discusses your options and gives a rundown of the pros/cons of each.

Michael Petrino, vice president at PTS Data Center Solutions, also appears in this issue of Processor. In Bruce Gain’s article, “Data Center Moving Day: There Is No Such Thing As ‘Over Planning’”, Michael shares his thoughts on how to prepare for a data center relocation project. Topics covered include the overall planning process, what to look for when hiring professional movers, the costs of up-time and down-time, transport options, and other complications.

Click on the links above to read the articles, or view the entire issue as a PDF.

Data Center Energy Summit 2008

On June 26th, the Silicon Valley Leadership Group (SVLG) held its first Data Center Energy Summit in Santa Clara, CA. The industry event focused on issues involving data center sustainability, energy efficiency and green computing.

In conjunction with Accenture and the Lawrence Berkeley National Laboratory (LBNL), the SVLG also unveiled a report containing real world case studies from its Energy Efficient Data Center Demonstration Project. You can download the report here: http://accenture.com/SVLGreport. Put together in response to the Environmental Protection Agency (EPA)’s report to Congress on data center energy efficiency, the report examines a number of innovative energy-saving initiatives.

Ken Oestreich from Cassatt points out in his blog that the bulk of the projects focused on improving infrastructure. He raises the following point:

My take is that the industry is addressing the things it knows and feels comfortable with: wires, pipes, ducts, water, freon, etc. Indeed, these are the ‘low-hanging fruit’ of opportunities to reduce data center power. But why aren't IT equipment vendors addressing the other side of the problem: Compute equipment and how it's operated?

I agree with Oestreich that methods for reducing the energy consumption of IT equipment definitely need to be explored further, but I think this report is a great step forward for the industry in terms of validating the EPA’s research and providing actionable data. I’m sure we’ll see more regarding IT equipment operations in future research.

As a side note, Data Center Knowledge has set up a calendar to help data center professionals keep track of upcoming industry events. Check it out: DataCenterConferences.com.

PTS Data Center Solutions Turns 10-Years Old!

Time flies like the wind. Fruit flies like bananas.
-- Groucho Marx

All kidding aside, time really does fly! It's hard for me to believe, but it was a decade ago that we founded PTS Data Center Solutions (known way-back-when as Power Technology Sales, Inc.). Our goal then, as it is now, was to provide our clients with unparalleled service and optimal solutions to meet their data center and computer room needs.

As we celebrate the company’s tenth anniversary, I’d like to express my appreciation to our hardworking team of consultants, engineers, designers, field service technicians, IT personnel and business staff, as well as our families, friends, business colleagues and clients for being part of our success.

While founded and headquartered in Franklin Lakes, New Jersey, our firm has experienced significant growth over the years, starting with the opening of our West Coast office in Orange County, California in 2004. Just a few years later, PTS Data Center Solutions completed the expansion and reorganization of our NJ facilities – an accomplishment that doubled the amount of useable office and warehouse space available to our team. We also upgraded our computer room, which hosts PTS' live environment and operates as a demonstration center for potential clients to see our work first-hand.

Over the course of the last decade, we’ve had the pleasure of working with small and medium-sized companies as well as large enterprise organizations across a broad spectrum of industry verticals. We’ve grown to become a multi-faceted turnkey solutions provider, offering services for consulting, engineering, design, maintenance, construction, monitoring and more. One of the more recent additions to our business offerings is our Computational Fluid Dynamic (CFD) Services, which use powerful 3-D CFD software for the design, operational analysis, and maintenance for data center and computer rooms of all types and sizes.

Our online presence has also grown. We’ve expanded our corporate website several times to provide new resources for our visitors. To help provide our clients and other IT professionals with insights on common data center issues, we began blogging in 2006. (I’d like to thank all of our readers for your comments and ongoing support!) Just a few months ago, we launched our own Facebook Page to help you stay up-to-date with the latest blog posts, our speaking engagements and other upcoming events. And, coming soon, look for me to be a guest blogger for the “World’s Worst Data Centers” contest, sponsored by TechTarget and APC.

This really is an exciting time for everyone at PTS Data Center Solutions. Reaching this milestone is a great achievement for our company and we’re looking forward to what the next ten years have to offer. Here's to the decades ahead!

Article: “Changing The Oil In Your Data Center”

This is just a quick update before everyone heads out for the holiday weekend.

If you haven't already done so, I encourage you to check out the May 16, 2008 issue of Processor magazine (Vol.30, Issue 20). Drew Robb interviewed me for his latest article, entitled “Changing The Oil In Your Data Center.”

Maintenance neglect is an all-too-frequent cause of unplanned data center downtime. This people-related problem stems from improper documentation of the maintenance process, failure to adhere to a set maintenance schedule, and the overlooking of critical systems. In the article, Robb talks with me about the value of implementing a scheduled maintenance plan to ensure reliable data center operations.

He also includes insights from several other data center professionals, including Steven Harris, director of data center planning at Forsythe Solutions Group (www.forsythe.com), and James Rankin, a CDW technology specialist (www.cdw.com). To read the full article, please visit the Processor website at http://www.processor.com/).

Have a safe and happy Memorial Day weekend!

The National Data Center Energy Efficiency Information Program

Matt Stansberry, editor of SearchDataCenter.com, posted a blog entry on the National Data Center Energy Efficiency Information Program. I'm echoing his post here because energy efficiency is such a critical issue for the industry.

The U.S. Department of Energy (DOE) and U.S. Environmental Protection Agency (EPA) have teamed up on a project with aims to help reduce energy consumption in data centers. In addition to providing information and resources which promote energy efficiency, the National Data Center Energy Efficiency Information Program is reaching out to data center operators and owners to collect data on total energy use.

In the words of the EPA’s Andrew Fanara:

We've put out an information request to anyone who has a data center to ask if you would measure the energy consumption of your data center in a standardized way and provide that to us. That will help us get a better handle on what's going on nationally in terms of data center energy consumption.

Hear the EPA’s Andrew Fanara talk about the program in this video from the Uptime Institute Symposium:



If you’d like to get your data center involved, more information can be found at the EPA's ENERGY STAR data center website and the DOE's Save Energy Now data center website.

CFD Modeling for Data Center Cooling

Computational fluid dynamics (CFD) modeling is a valuable tool for understanding the movement of air through a data center, particularly as air-cooling infrastructure grows more complex. By using CFD analysis to eliminate hot spots, companies can lower energy consumption and reduce data center cooling costs.

Mark Fontecchio at SearchDataCenter.com has written a great new article on the subject, entitled “Can CFD modeling save your data center?”. Fontecchio examines the use of CFD analysis as a tool for analyzing both internal and external data center airflow.

In the article, Carl Pappalardo, IT systems engineer for Northeast Utilities, provides a first-hand account of how CFD analysis helped in optimizing their data center’s cooling efficiency. Allan Warn, a data center manager at ABN AMRO bank, also shares his thoughts on the value of renting vs. buying CFD modeling software. Fontecchio also includes insights from industry experts, including Ernesto Ferrer, a CFD engineer at Hewlett-Packard Co., and from yours truly, Pete Sacco.

For more information on data center cooling, download my White Paper, “Data Center Cooling Best Practices”, at http://www.pts-media.com (PDF format). You can also download additional publications, like Vendor White Papers, from the PTS Media Library.

To learn more about how PTS uses CFD modeling in the data center design process, please visit: http://www.ptsdcs.com/cfdservices.asp.

Free White Paper on Relative Sensitivity Fire Detection Systems

Fire detection is a challenge in high-end, mission critical facilities with high-density cooling requirements. This is due primarily to the varying levels of effectiveness of competing detection systems in high-velocity airflow computer room environments.

In a new white paper, PTS Data Center Solutions’ engineers Suresh Soundararaj and David Admirand, P.E. identify and analyze the effectiveness of relative sensitivity-based fire detection systems in a computer room utilizing a high-density, high-velocity, and high-volume cooling system.

In addition to examining the differences between fixed sensitivity and relative sensitivity smoke detection methodologies, Soundararaj and Admirand detail the results of fire detection tests conducted in PTS’ operational computer room and demo center using AirSense Technology’s Stratos-Micra 25® aspirating smoke detector.

The illustrated 13-page white paper, entitled “Relative Sensitivity-based Fire Detection Systems used in High Density Computer Rooms with In-Row Air Conditioning Units,” is available for download on our website in PDF format.