Data Center Network
sponsored by Dell Force10
Improving Energy Conservation in Your Data Center
By David SimsYou want to go green with your data center energy practices? That’s great. Conservation is good. Green is good. But don’t think you’re doing it just for the benefit of Mother Nature; according to Kate Brew, Product Marketing Manager from Anue Systems, you’re doing it for efficiency and good, hard ROI numbers.
And that includes everybody connected with your IT facility, since as she says they all pretty much have to be on the same page if you’re going to succeed.
Brew cites a report released in late 2011 by Forrester (News - Alert), titled “Power and Cooling Heat Up the Data Center,” finding that you’re spending 70 percent of your operations costs for “an average data center facility” on energy.
Not only that, as the Forrester report said, the fact that you have to put the networking appliances so far apart to let the heat dissipate means that you’re wasting lots of rack space. The report writers say they’ve seen data centers where fully two-thirds of the rack space is just wasted to allow for cooling.
It’s a tricky area to get right. Last June TMC’s (News - Alert) Beecher Tuttle noted that Michael Petrino, vice president for PTS Data Center Solutions, discussed software providing redundancy planning and 3D graphics “for space, power and cooling capacity.”
According to Tuttle, Petrino told Processor (News - Alert).com that "With an accurate room model, we are able to take information regarding new systems and run multiple room and rack cooling capacity and redundancy planning scenarios as to where in the room the heat generated can be most effectively dissipated without impacting other systems.”
Brew writes that networking chews up about ten percent of your data center’s power consumption, and figures to become an even bigger slice of your data center energy consumption with the increased efficiency we’re seeing in servers and storage. So if you’ve been able to get away without paying much attention to the energy and space you’re devoting to them so far, but that might need to change.
She adds that network monitoring switches, with their de-duplication capabilities, can be used with great effect in this. Finding smaller and more efficient switches is key, and Brew offers three specs you can use to help in this search:
First, check the density. How many network ports and tools can be configured for it?
Second, determine the rack space. How much actual space is needed for the app, without configuration.
Third, of course, look at the power consumption. The important thing here is the watts per port; Brew says it should be as low as possible. Check how much power consumption there will be per port, since that’s where your real rack heat dissipation and energy usage figures will come from.
David Sims is a contributing editor for TMCnet. To read more of David’s articles, please visit his columnist page. He also blogs for TMCnet here.
Edited by Rich Steeves
Data Center Network White Papers
Distributed Core Architecture Using the Z9000 Core Switching System
As data centers scale to support thousands of servers, IT managers are seeking better ways to network those servers while reducing costs and power consumption. Moreover, in large-scale data center cluster environments inter-node communication bandwidth is increasingly becoming the main bottleneck. For these environments, applications need to exchange information with remote nodes for execution of their local computation. For example, web search engines require parallel communication with every node in the cluster to provide the most relevant results, and web servers may require interaction with hundreds of sub-services running on remote nodes.Read the full white paper
Will Traffic Spikes Overwhelm Your Data Center Network?
One of the most significant data center technologies introduced over the last ten years is virtualization. Both server and storage virtualization are commonly found in today’s data centers, bringing more flexibility, greater efficiency and higher availability to IT computing environments. As we try to leverage server and storage virtualization, we need to look at how we deploy network virtualization judiciously so as to implement an optimized data center. As virtualization technology expands beyond the test and development environments where it started to production environments, IT managers are learning the impact of virtualization on the network. With most virtualized servers supporting between 2 and 10 applications, the network traffic profile is changing. Not only is network traffic increasing, but the potential for very large network traffic spikes has increased as well. With each of these virtualized servers, each rack now hosts between 160 and 400 application instances. IT managers must take this into consideration as they deploy more virtualized servers in their data centers.Read the full white paper
Network Automation with the Dell Force10 Open Automation Framework
It has been said that the shift to a virtual data center will be the most significant IT transformation since the invention of the mainframe. This transformation promises to join virtualized computing and storage layers to a virtualized network stack, while enabling new agile models such as cloud computing. At the same time, it is vital for the success of this technology that the transition to the virtual data center not bring back the proprietary restrictions and limitations that were the hallmark of the mainframe era.Read the full white paper
