August 18, 2009 1:22 PM PDT
Energy-aware Internet routing coming soon
by Marguerite Reardon 7 comments
Researchers have come up with a new way to route Internet traffic that could save big Internet companies like Google millions on their electricity bills, according to an article published by MIT's Technology Review.
Researchers from MIT, Carnegie Mellon University, and the networking company Akamai recently published results from a study that suggest big Internet companies could save up to 40 percent on their electricity bills by using an algorithm to send Internet traffic to data centers where electricity is less expensive.
Data centers consume a lot of energy, which costs operators like Google and Amazon millions of dollars to run each year. And now as more digital information is "virtualized" and accessed in the cloud, centralized data centers are getting even bigger and are consuming even more energy.
In fact, data center energy usage is expected to quadruple during the next decade in the absence of efforts to improve efficiency, according to the article, which referenced a report from McKinsey & Company and the Uptime Institute published in July 2008. The article also estimated that today, some large Internet companies spend more than $30 million a year on energy costs.
But now there could be a way for these companies to lower their energy costs. Using an algorithm developed and tested by researchers at MIT, Carnegie Mellon, and Akamai, these companies can track electricity prices that fluctuate by location and time of day. It then calculates the best option depending on the price of electricity and the distance that data must travel to get to a particular location, to provide the most cost-effective route for data based on energy costs.
There is mounting pressure for big Internet companies to reduce their energy usage. Not only is it expensive, but these companies face pressure from governments and others concerned with the environment to reduce their carbon footprints.
The algorithm developed by these researchers doesn't necessarily cut the use of energy. It merely helps companies better manage their energy costs. But researchers argue it could also be used to direct traffic to facilities where greener forms of energy are being used.
The example used in the article is a Google data center in Belgium that relies on ambient cooling to keep its facilities under a certain temperature instead of firing up expensive air-conditioning systems. On days when the weather is too warm, the servers in the data center simply shut down. One researcher working on the project said that the energy-routing algorithm is an extension of this idea.
Managing energy and its costs is likely to be big business in the coming years for technology companies. Cisco Systems, the largest networking company in the world, recently said it sees a $100 billion market in helping power utilities better manage their electrical grids using communication technology.
Other high-tech companies, including IBM, Intel, and several start-ups, are also ramping up to develop sensors and other technology that alert power companies and their customers about energy usage in an effort to balance usage and reduce costs.
The idea of the "smart grid" is to modernize the electricity industry by overlaying digital communications onto the grid. For example, smart meters in a person's home can communicate energy usage to utilities, allowing these companies to more efficiently manage the electricity supply and potentially allow a consumer to take advantage of cheaper rates.
Some utilities are already testing technology from Cisco that will help its routers and switches more efficiently manage the flow of electricity and prevent outages in grid distribution equipment. Cisco also released new software earlier this year called EnergyWise, which allows more efficient use of energy in office buildings. The software is a free upgrade to Cisco Catalyst switches that can monitor and manage how energy is used on IP-connected devices, including phones and wireless routers. Using the software, companies can set policies on energy use, allowing PCs or networking equipment to go into sleep mode after work hours, for example
Showing posts with label Akamai. Show all posts
Showing posts with label Akamai. Show all posts
Tuesday, September 29, 2009
Friday, August 28, 2009
Energy-Aware Internet Routing
Energy-Aware Internet Routing
Software that tracks electricity prices could slash energy costs for big online businesses.
By Will Knight
Monday, August 17, 2009
An Internet-routing algorithm that tracks electricity price fluctuations could save data-hungry companies such as Google, Microsoft, and Amazon millions of dollars each year in electricity costs. A study from researchers at MIT, Carnegie Mellon University, and the networking company Akamai suggests that such Internet businesses could reduce their energy use by as much as 40 percent by rerouting data to locations where electricity prices are lowest on a particular day.
Data beast: Google maintains a huge datacenter in The Dalles, OR.
Credit: John Nelson
Modern datacenters gobble up huge amounts of electricity and usage is increasing at a rapid pace. Energy consumption has accelerated as applications move from desktop computers to the Internet and as information gets transferred from ordinary computers to distributed "cloud" computing services. For the world's biggest information-technology firms, this means spending upwards of $30 million on electricity every year, by modest estimates.
Asfandyar Qureshi, a PhD student at MIT, first outlined the idea of a smart routing algorithm that would track electricity prices to reduce costs in a paper presented in October 2008. This year, Qureshi and colleagues approached researchers at Akamai to obtain the real-world routing data needed to test the idea. Akamai's distributed servers cache information on behalf of many large Web sites across the US and abroad, and process some 275 billion requests per day; while the company does not require many large datacenters itself, its traffic data provides a way to model the demand placed on large Internet companies.
The researchers first analyzed 39 months of electricity price data collected for 29 major US cities. Energy prices fluctuate for a variety of reasons, including seasonal changes in supply, fuel price hikes, and changes in consumer demand, and the researchers saw a surprising amount of volatility, even among geographically close locations.
"The thing that surprised me most was that there was no one place that was always cheapest," says Bruce Maggs, vice president of research at Akamai, who contributed to the project while working as a professor at Carnegie Mellon and is currently a professor at Duke University. "There are large fluctuations on a short timescale."
The team then devised a routing scheme designed to take advantage of daily and hourly fluctuations in electricity costs across the country. The resulting algorithm weighs up the physical distance needed to route information--because it's more expensive to move data further--against the likely cost savings from reduced energy use. Data collected from nine Akamai servers, covering 24 days of activity, provided a way to test the routing scheme using real-world data. The team found that, in the best scenario--one in which energy use is proportional to computing--a company could slash its energy consumption by 40 percent. "The results were pretty surprising," Maggs says.
The ability to throttle back energy consumption could have another benefit for massive Internet companies, the researchers say. If an energy company were struggling to meet demand, it could negotiate for computation to be moved elsewhere; the researchers say that the market mechanisms needed to make this possible are already in place.
[1] 2 Next »
Tags Akami electric grid electricity Google Microsoft
Related Articles
»Netbook Chips Create a Low-Power Cloud
04/16/2009 »A Picowatt Processor
07/08/2008 »Saving Energy in Data Centers
03/11/2008 Comments Close Comments
Open 1 Comment
Price Standards Needed
Writing this kind of software and many other kinds of software to manage usage is pretty straightforward-- really, no fancy routers needed. The only hard part is just getting the price data! Most likely, a complex perl script would be required for each utility to scrape data off a web page-- if it's even available. It could be that some special equipment would be required to receive and decode price data-- then it's very difficult to even get data.
The FERC should solicit proposals for a simple standard format to post current and predicted prices for power via WWW (or even email/NNTP). Ideally, a simple format would be used as the early internet RFCs did. I suppose XML could be used, though I think a simpler text format would be better. Coming up with a generic way to represent "smart-grid" status, and then requiring the information to be posted by all utilities the same way is the main step needed to make grids smarter. Otherwise, we could end up with the Grid of Babble, where each utility uses proprietary equipment, networks, and protocols, and innovation is hobbled.
Software that tracks electricity prices could slash energy costs for big online businesses.
By Will Knight
Monday, August 17, 2009
An Internet-routing algorithm that tracks electricity price fluctuations could save data-hungry companies such as Google, Microsoft, and Amazon millions of dollars each year in electricity costs. A study from researchers at MIT, Carnegie Mellon University, and the networking company Akamai suggests that such Internet businesses could reduce their energy use by as much as 40 percent by rerouting data to locations where electricity prices are lowest on a particular day.
Data beast: Google maintains a huge datacenter in The Dalles, OR.
Credit: John Nelson
Modern datacenters gobble up huge amounts of electricity and usage is increasing at a rapid pace. Energy consumption has accelerated as applications move from desktop computers to the Internet and as information gets transferred from ordinary computers to distributed "cloud" computing services. For the world's biggest information-technology firms, this means spending upwards of $30 million on electricity every year, by modest estimates.
Asfandyar Qureshi, a PhD student at MIT, first outlined the idea of a smart routing algorithm that would track electricity prices to reduce costs in a paper presented in October 2008. This year, Qureshi and colleagues approached researchers at Akamai to obtain the real-world routing data needed to test the idea. Akamai's distributed servers cache information on behalf of many large Web sites across the US and abroad, and process some 275 billion requests per day; while the company does not require many large datacenters itself, its traffic data provides a way to model the demand placed on large Internet companies.
The researchers first analyzed 39 months of electricity price data collected for 29 major US cities. Energy prices fluctuate for a variety of reasons, including seasonal changes in supply, fuel price hikes, and changes in consumer demand, and the researchers saw a surprising amount of volatility, even among geographically close locations.
"The thing that surprised me most was that there was no one place that was always cheapest," says Bruce Maggs, vice president of research at Akamai, who contributed to the project while working as a professor at Carnegie Mellon and is currently a professor at Duke University. "There are large fluctuations on a short timescale."
The team then devised a routing scheme designed to take advantage of daily and hourly fluctuations in electricity costs across the country. The resulting algorithm weighs up the physical distance needed to route information--because it's more expensive to move data further--against the likely cost savings from reduced energy use. Data collected from nine Akamai servers, covering 24 days of activity, provided a way to test the routing scheme using real-world data. The team found that, in the best scenario--one in which energy use is proportional to computing--a company could slash its energy consumption by 40 percent. "The results were pretty surprising," Maggs says.
The ability to throttle back energy consumption could have another benefit for massive Internet companies, the researchers say. If an energy company were struggling to meet demand, it could negotiate for computation to be moved elsewhere; the researchers say that the market mechanisms needed to make this possible are already in place.
[1] 2 Next »
Tags Akami electric grid electricity Google Microsoft
Related Articles
»Netbook Chips Create a Low-Power Cloud
04/16/2009 »A Picowatt Processor
07/08/2008 »Saving Energy in Data Centers
03/11/2008 Comments Close Comments
Open 1 Comment
Price Standards Needed
Writing this kind of software and many other kinds of software to manage usage is pretty straightforward-- really, no fancy routers needed. The only hard part is just getting the price data! Most likely, a complex perl script would be required for each utility to scrape data off a web page-- if it's even available. It could be that some special equipment would be required to receive and decode price data-- then it's very difficult to even get data.
The FERC should solicit proposals for a simple standard format to post current and predicted prices for power via WWW (or even email/NNTP). Ideally, a simple format would be used as the early internet RFCs did. I suppose XML could be used, though I think a simpler text format would be better. Coming up with a generic way to represent "smart-grid" status, and then requiring the information to be posted by all utilities the same way is the main step needed to make grids smarter. Otherwise, we could end up with the Grid of Babble, where each utility uses proprietary equipment, networks, and protocols, and innovation is hobbled.
Subscribe to:
Posts (Atom)