Raising the temperature set point in data centers becomes a key objective for IT managers following the updated ASHRAE 2008 environmental recommendations for IT equipment. Running server rooms at higher supply temperatures improves energy efficiency, reduces costs and opens paths for new data center designs (i.e., using economizers and “free” cooling).

Temperature set point increases call for improved monitoring, real time hot-spot detection and alerting through the deployment of thermal and air flow sensors on data center equipment. This differs radically from current practices of environmental auditing done once or twice a year using thermal cameras.

The ability for IT operators to quickly react to an abnormal thermal event can be the key difference between overcooling (and incurring higher energy bills) and adopting new set points in a well-managed facility where “hot spot” impacts are avoided.

This calls for the adoption of energy and environmental optimization solutions that can:

• Report all real-time temperature, humidity and air flow sensor data
• Generate alerts when defined thresholds are reached
• Display on 2D and 3D maps the temperature recorded at various levels such as 1U, 21U and 42U in a server rack

Various energy and environmental metrics, monitored in real time and over extended periods, can be of enormous help to data center operators in rapidly identifying familiar incidents that affect data center performance. For example, new tools available today can record the impact of a partial power outage on temperatures and power metrics.

Figure 1 shows a partial power failure and its direct impact on temperature change. Note that the partial power outage has led to a drop in power demand for the monitored electric load. Figure 2 shows how cold aisle (inlet) and hot aisle (exhaust) temperatures start equalizing due to a partial power failure of the cooling system. In Figure 3, the same root cause generates a quick temperature rise in the IT server room, generating threshold alerts.

Operators can leverage this correlation and analysis of metrics to adjust their thresholds and get quickly alerted to a set of events that they have identified as a partial power outage. When such events occur, they can immediately contact maintenance services to correct the root cause and/or properly shut down the servers, avoiding even more serious damage.

Figure-1: Temperature versus Power Outage

Figure-2: Watch the lab server inlet and exhaust temperatures equalize

Figure-3: Watch the IT server room temperatures rise

Electric utility-funded incentive programs such as the 80 PLUS program in North America certify – and pay incentives for – computers and data center servers that utilize energy-efficient power supplies.

The 80 PLUS performance specification requires that multi-output power supplies in computers and servers be 80% or greater energy-efficient at 20%, 50% and 100% of rated load with a true power factor of 0.9 or greater.

By monitoring power factor and load metrics with newly available tools (see figure below), IT managers get a real-time view of energy efficiency for their equipment’s power supplies. This makes it easier to detect abnormal behaviors and plan for preventive maintenance when a power supply moves out of its specification’s range.

Whatever their global location, IT managers soon may have to self-declare or demonstrate their data center’s compliancy with a regional “Ecolabel” for energy efficiency. For example, the U.S. Energy Star program for data centers or the EU Code of Conduct for Data Centres initiatives have published technical specifications that call for endorsement of best practices. In the case of Energy Star, data center operators entering their information in the U.S. Environmental Protection Agency (EPA) database will get a 1-to-100 rating. A score of 50 indicates average performance; 75 or higher means the facility ranks in the top 25 percent in terms of energy efficiency, qualifying for an Energy Star label.

Seeking certification of an organization’s Power Usage Efficiency (PUE) by The Green Grid is a step in the right direction. But because data centers are so dynamic in nature, tracking energy consumption only when trying to pass a certification requirement doesn’t show the full picture. Without historical data, it may even be difficult to understand why a rating has changed. This points up the need for new operational tools to deliver real-time Ecolabel assessments of the data center, as shown in the figure below.

Follow us here.

We’ve just launched a new product, the Energy Optimizer, that helps commercial facilities managers answer that question.

The EO takes you from a once-a-month utility bill to a real-time web application that lets you break down energy usage by building area and purpose, set budgets, and export all the data through our open APIs for further analysis in other tools.

Now, you can make better decisions about facilities upgrades and policy changes, measure the effect of the changes, and get ongoing verification that your changes are reducing energy usage and helping the bottom line.

Check it out: Arch Rock Energy Optimizer

PhyNet 3.5 breaks ground for a new industry of environmental, municipal, agricultural, and energy experts who monitor and manage remote multi-client sites as a real-time service. By enhancing the PhyNet tiered architecture with support for IPSec (Internet Protocol Security) and NAT (Network Address Translation) traversal, new cloud-based services can be introduced that pass data securely through corporate Internet connection boundaries.

PhyNet 3.5 also provides a number of new service provider-oriented features for more efficient deployment, including plug-and-play installation, location finding nodes, and remote management and diagnostics. It is available free as a download for our installed-base customers (restrictions apply) and has begun shipping with all products.

http://www.archrock.com/products/

Time’s Best Inventions of 2008 list has “The Internet of Things” at #30!

In September, a group of high-tech companies that includes Cisco and Sun formed the IP for Smart Objects Alliance. Simply put, the organization intends to create a new kind of network that will allow sensor-enabled physical objects — appliances in your home, products in a factory, cars in a city — to talk to one another, the same way people communicate over the Internet.

I’m happy to see these ideas hitting a broader audience. When your relatives ask you what you’re interested in, you can send them this article.

Another example: if you happen to have a subscription to The Economist (still pretty high up on the mainstream scale), you can find a more in-depth special report from last year called When everything connects:

In coming years wireless will vanish entirely from view, as communications chips are embedded in a host of everyday objects. Such chips, and the networks that link them together, could yet prove to be the most potent wireless of them all.

What could we do if we combined two proven technologies in our embedded networks: the scalable and link-independent IP network layer plus the agreed-upon device definitions and security model of the ZigBee Application Profile system?

You could have thermostats and electric meters communicating directly with other devices like displays and inverters, regardless of the physical networking technology chosen by their manufacturers. 802.15.4 wireless devices could connect with power-line devices like HomePlug, WiFi devices, and Ethernet devices through a routed network without application-layer gateways that may need to be upgraded whenever you add new features.

You could also introduce distance into the picture, directly linking web servers and application servers with in-premise devices to give you better visibility and control over your energy usage or anything else that benefits from embedded networking.

If you’re interested, check out our page on the CAP proposal, which suggests a technique for adapting the ZigBee Application layer to run over a native IP network. There’s a set of overview slides on that page.

More technically minded folks might want to read the CAP Internet-Draft.

I’ll kick off Arch Rock’s new blog with some industry news.

Several companies (including Cisco, Duke Energy, Emerson, and of course, Arch Rock) recently joined together to promote IP (Internet Protocol) for Smart Objects (IPSO).

What’s IP? It’s the same networking protocol everything else runs — your server, your laptop, your router, your mobile phone, your printer, your VOIP phone, and your building management system.

What’s a smart object? It’s a thing that doesn’t yet have a network connection but has tons of valuable data to exchange with us. Things like parking meters, thermostats, water meters, and light switches. They’re tiny. They often run on batteries. They use sensors to connect with the physical world, just like the devices above connect with people. And if we can connect with them, we can be greener, more efficient, and more effective.

So why IP for smart objects? Because IP already runs everything else. It’s the most ubiquitous, most versatile networking protocol we have, and since there will be even more smart objects than there are currently nodes on the Internet, we need the most scalable protocol we can find. No gateways needed. Just simple, stable, manageable IP connectivity. To everything.

Check out the IPSO site for more details.

And, check back here for news about Arch Rock and for ideas drawn from the wider worlds of smart objects and embedded networks.