Ads by TechWords

See your link here
Receive the latest technology news and information.
Computerworld Daily News (First Look and Wrap-Up)
Computerworld Blogs Newsletter
The Weekly Top 10
Cloud Computing
View all newsletters




Privacy Policy
 

Opinion: Green In and Out

September 3, 2007 12:00 PM ET

Computerworld - There are two kinds of people: optimists and pessimists. Sadly, Im one of the latter. So I wasnt surprised when the vast majority of scientists concluded that human contributions to the buildup of greenhouse gases are a key component of global warming.

I sometimes wonder, Whats the use? After all, China and India are just getting started as major producers of greenhouse gases, and the U.S. government seems unwilling to lead this nation in efforts to conserve energy. The future looks very bleak.

Unless you talk to the right optimists.

I recently had the privilege of chatting with John Engates, chief technology officer at Rackspace Managed Hosting in San Antonio. His company has eight global data centers running more than 30,000 servers for 10,000 customers, primarily small and midsize companies. Engates is not only an optimist, but hes also someone with the authority to make changes in how his company consumes energy within its core business operations  the data center.

Naturally, Rackspace is already doing a lot of the things that experts say can cut energy consumption in the data center (see "Seven Steps to a Green Data Center," June 25, 2007). Like many organizations, the company is aggressively pursuing server virtualization. The goal is not only to reduce its server count, but also to get better utilization of system memory, Engates says. As Robert Mitchell pointed out (see "Memory: The New Power Hog," April 30, 2007), RAM power requirements in new servers are double those of CPUs.

As data center managers seek to cut power costs, servers get most of the focus. But Rackspace looks at all the gear it uses. Energy concerns now affect how the company buys and deploys network switches and routers. Engates goal is to reduce unused network ports, which translates into fewer machines and less power used.

Engates emphasizes that thinking green goes beyond looking at what you plug into an electrical socket. You need to look at your entire operation. Because Rackspace buys so many identical servers from the same vendor, it has arranged for the supplier to stop including redundant cables and manuals in each box, so theres less packaging to throw away.

At a more strategic level, Rackspace has committed to using electricity generated from renewable resources. Its new data center in the U.K. will be run completely by biomass- generated power. Its electricity that will be grown instead of mined, Engates says.

For each server Rackspace buys and places in U.S. data centers where renewable energy sources arent available, the company buys offset energy credits from Native Energy LLC in Charlotte, Vt. Native Energy uses that money to develop renewable-energy projects. And Rackspace has joined The Green Grid, a consortium dedicated to cutting data center power consumption.



Jump to comments

Mark Hall

Additional Resources

Microsoft
Here are some of the key reasons why you would want to run Unified Access Gateway with DirectAccess.
Microsoft
Review how one energy firm tightened protection and simplified IT work using business-ready security solutions.
Sybase
In this white paper, IDC analyzes the role of next-generation mobile enterprise platforms as organizations seek a more strategic deployment of mobile solutions.

Learn the important issues you must consider before starting your next mobility initiative. Get your mobility white paper from IDC now, compliments of Sybase.

What People Are Saying

White Papers & Webcasts

5 Steps to Faster Data Classification
Find Sensitive Data More Quickly. Read More Now.  

Maximize ROI for Web Applications
Register for this webcast now!

Effectively Implementing Datacenter Automation
Effectively select and deploy the best datacenter automation solution today!

Mitigate Risk, Lower Costs and Improve Network Efficiency
Create a stable IP network that not only meets today's challenges, but is flexible enough to also meet future demands.

Optimize Performance of Datacenter to Datacenter Traffic
To get the backups and database synchronizations completed on time, enterprises rely on WAN optimization from Blue Coat.  

Three IT Strategies to Cut Cost Intelligently
Register for this Webcast! Provided by BMC Software.

Delivering IT to the Virtual Workforce
Learn how to meet 4 key virtual workforce needs.  

Interactive Guide: Getting Started with Data Governance
Download this Interactive Guide today!


IT Jobs

 


Enterprise Data Center Network Reference Architecture
Simply designing a data center that only deploys more servers, storage and devices significantly increases network complexity and cost. The data center network also must offer components such as security, performance acceleration, high density and a resilient network infrastructure. This document shares Juniper Networks best practices in designing a highly efficient, secure, scalable and flexible data center network.

View this webcast 
Advancing the Economics of Networking
Aging network systems and old habits have dictated how businesses spend their IT budgets. As a result, a large percentage of IT dollars are being spent to “stay in the race.” While this model keeps revenue streams flowing for legacy network vendors, it doesn’t necessarily help businesses gain a competitive advantage. Juniper is changing this economic model by delivering a new family of solutions that reduce capital and operational expenses. By freeing up IT budget dollars, Juniper allows businesses to invest in other innovative technologies that will further reduce the cost of doing business while improving their competitive position.

Download this white paper 
Implementing HA at the Enterprise Data Center Edge to Connect to a Large Number of Branch Offices
This paper reviews the problem of creating a network where the dynamic availability of services is of critical importance. It explains how to deploy a highly available, working cluster by taking the design considerations in this paper and applying them to an actual deployment. The subsequent sections include configuration bits, diagrams and detailed discussions of how to accomplish a High Availability (HA) data center deployment.
Download this white paper