Big data: polluters or environmentalists

The problem, as the New York Times pointed out in a previous article in its publication: All of the hardware is being powered up and running, even including industrial cooling and standby diesel generators emitted into the atmosphere and other pollutants. At the same time, even if data center operators continue to reward those energy-saving measures, continuous uptime over its energy efficiency is often running at full power, but there are still many servers apparently idle or apparently underutilized.

As the volume of data continues to grow, there will also be a significant amount of pollution from the data center. This, as the New York Times has sharply reported, is clearly inconsistent with the image of the information industry as fashionable, efficient and environmentally friendly.

I do not fully agree with this assessment. Yes, the data center of the cloud computing era is obviously more polluting than a normal family or even a big city apartment building. But overall, even as the New York Times approves, all data centers consume only 2% of their energy resources. And we have a long way to go, and the data center industry has made significant progress in improving the efficiency of its data centers.

To more specific http://www.aliyun.com/zixun/aggregation/14294.html "> 's large data, hardware efficiency is improving by adopting a series of measures such as the deployment of Hadoop. According to a study reported last month. One of these is VMware's newest open source Serengeti project, enabling administrators to deploy Hadoop virtualization on existing hardware. This increases efficiency while avoiding the need to buy new hardware.

Another project is the Datarush product of pervasive software:

Complex parallel Hadoop operations abstracted by Datarush and Rushanalyzer enable users to monitor I/O and CPU usage in real time and reduce memory limitations. Ultimately, the server can use up to 80% utilization, and can be more rapid and iterative analysis of large data.

Another point that the New York Times ignores is that most data-analysis experts are supported by these data centers for further sustainable research on climate change and the environment. For example, the Climate Simulation Center established by NASA has built a customized large data platform that collects up to a few PB of meteorological data to better predict and understand global climate change. The NCCS data center consumes a lot of power, but the researchers found through large data analyses what could lead to limiting or even reversing global warming. I want to say that this is just a fair deal.

Of course, not every big Data project aims to promote greater benefits. But there are some such projects in the climate science and health and education industries.

I applaud the New York Times for focused discussions on this issue. But let's keep this in perspective. Large data does emit a large amount of pollutants in the atmosphere, but the large data industry is also developing more efficient computing methods and addressing the challenges of some of the toughest environments where humans will likely make up for any of these data center operations, more efficient calculations that will directly affect the atmospheric environment.

(Responsible editor: Schpeppen)