The most popular and comprehensive Open Source ECM platform
Let us know more
Green Clouds: The Efficiencies of Cloud Computing
Cloud computing has the potential to be more environmentally friendly than existing data centers. Doug Washburn and Lauren Nelson, analysts at Forrester Research, recently identified at least three ways that cloud computing can be considered as more ‘green’ than traditional data centers:
- Lower Energy Costs
- Fewer Carbon Emissions
- Reduced Electronic Waste
- More Efficient Server Utilization — Use only what you need, when you need it
By using server virtualization, cloud computing centers can consolidate into a single location servers that are highly shared and well utilized. The result is that fewer total servers are needed. In additional, cloud computing providers, whose main business is to efficiently manage their data center, are highly motivated to seek out and deploy efficiencies for achieving cost savings.
Traditional data centers located within and managed by a single organization tend to often still use out-dated server machines with poor energy efficiencies. In fact, eWeek reports that in the UK as many as 40 percent of traditional data centers are using old equipment with Power Usage Effectiveness (PUE) ratings ranging from 3-7, a measure that means that up to 7 watts of electricity are wasted for every one watt of electricity is actually used by servers for computing. Most newer data centers have PUE ratings closer to 1.2.
But the Forrester report suggests that cloud computing data centers can still do more. The emphasis on ‘green’ by cloud computing data centers to date has focused almost exclusively on energy savings. There are other areas where cloud data centers can be more ‘green’. In particular, the report identifies the management of e-waste as one area where cloud computing data centers can be more efficient. The report also suggests that companies with portfolios of data centers can also more efficiently measure and manage energy use across the different locations to determine when and where computing jobs can be most efficiently run.
