巨大的数据中心效率
提高可避免能源激增
Huge data center efficiency gains stave off energy surge
计算量增长了六倍，导致能源需求仅增长了6％
A sixfold increase in compute led to energy demands rising just six percent
February 27, 2020
By Sebastian Moss
本文由DCD中国授权翻译并在DeepKnowledge平台发表。
研究人员发现，尽管2010年至2018年间数据中心的计算负荷大幅增加，但实际能源使用量仅略有增加。
Despite a huge increase in the compute loads of data centers between 2010 and 2018, actual energy usage rose only marginally, researchers found.
效率的显着提高意味着该行业的电源需求在计算跃升550％的时间内仅增长了6％。这项由西北大学，劳伦斯伯克利国家实验室和Koomey Analytics进行的研究定于2月28日发表在《科学》杂志上。
Significant efficiency advances mean that the industry's power demands rose just six percent in the time it took for compute to jump 550 percent. The study - by Northwestern University, Lawrence Berkeley National Laboratory, and Koomey Analytics - is set to be published in the journal Science on February 28.
最后的好消息
Good news at last
研究人员使用来自各种来源的数据来构建数据中心能耗的图景，包括有关数据中心设备库存，效率趋势和市场结构的信息。生成的模型可以详细分析数据中心设备（例如服务器，存储设备和冷却系统），数据中心的类型（包括超大规模和云计算）以及按世界区域使用的能源。
Researchers used data from various sources to build a picture of data center energy consumption, including information on data center equipment stocks, efficiency trends, and market structure. The resulting model enables a detailed analysis of the energy used by data center equipment (such as servers, storage devices, and cooling systems), by type of data center (including hyperscale and cloud) and by world region.
研究人员说，能源效率的提高超过了其他主要行业，这意味着数据中心被认为占全球能源使用量的百分之一。
The energy efficiency improvements outpaced any other major sector, the researchers said, and mean that data centers are thought to account for one percent of global energy usage.
该研究批评“经常被引用但过于简单的分析”，声称过去十年来全球数据中心的能源使用量增加了一倍，并且在未来十年内其能源使用量将增加三倍甚至四倍。研究指出：“但是，基于最近的服务需求增长指标进行的这种推断忽视了同时出现的强大的抵消性能源效率趋势。”
The study criticized 'several oft-cited yet simplistic analyses' that claim the energy used by the world’s data centers has doubled over the past decade and that their energy use will triple or even quadruple within the next decade. 'But such extrapolations based on recent service demand growth indicators overlook strong countervailing energy efficiency trends that have occurred in parallel,' the study notes.
该论文估计，2005年全球数据中心的能源使用量为153太瓦小时（TWh），到2010年为194 TWh，2018年为203 TWh。
The paper estimates that in 2005 the worldwide energy use of data centers was 153 terawatt-hours (TWh), 194 TWh by 2010, and 203 TWh in 2018.
但是在2010年至2018年之间，全球数据中心的工作负载和计算实例增加了六倍，数据中心的Internet协议（IP）流量增加了十倍以上，数据中心的存储容量估计增加了25倍。
But between 2010 and 2018, global data center workloads and compute instances have increased sixfold, data center Internet protocol (IP) traffic has increased by more than 10-fold, and data center storage capacity has increased by an estimated factor of 25.
研究人员展望未来，指出了预测未来的困难，但表示，诸如更高级别的服务器虚拟化，向低功耗存储设备的转移以及从小型，低效的数据中心到超大规模设施的结构性迁移等改进措施应确保“是一种足够的能源效率资源，可以吸收数据中心计算实例再增加一倍的情况，而与此同时，全球数据中心能源消耗的增加可忽略不计。”
Looking ahead, the researchers noted the difficulty of predicting the future, but said that improvements such as greater levels of server virtualization, shifts to low-power storage devices, and structural moves from small, inefficient data centers to hyperscale facilities should ensure that 'there is a sufficient energy efficiency resource to absorb the next doubling of data center compute instances that would occur in parallel with a negligible increase in global data center energy use.'
我们还有工作要做
We still have work to do
尽管有正面消息，但该研究的负责人埃里克·马萨涅（Eric Masanet）告诫道：“尽管数据中心在效率方面的历史进步令人瞩目，但我们的发现并不意味着IT业和政策制定者可以为之骄傲。”
Despite the positive news, the study's lead, Eric Masanet, cautioned: “While the historical efficiency progress made by data centers is remarkable, our findings do not mean that the IT industry and policymakers can rest on their laurels.”
“我们认为仍有足够的剩余效率潜力，可以持续数年。但是对数据的需求不断增长意味着，包括政策制定者，数据中心运营商，设备制造商和数据消费者在内的每个人都必须加大努力，以避免在此十年后期能源使用量可能急剧上升。”
“We think there is enough remaining efficiency potential to last several more years. But ever-growing demand for data means that everyone - including policymakers, data center operators, equipment manufacturers, and data consumers - must intensify efforts to avoid a possible sharp rise in energy use later this decade.”
LBNL研究科学家之一的合著者Arman Shehabi补充说：“考虑到数据中心是快速发展的行业中的能源密集型企业，我们确实需要对其进行严格的分析。较少的详细分析预测了数据中心能源使用的快速增长，但并没有完全考虑到行业在效率方面的历史进步。当我们把缺失的部分包括在内时，我们的数字生活方式就会出现不同的景象。”
Coauthor Arman Shehabi, LBNL research scientist, added: 'Considering that data centers are energy-intensive enterprises in a rapidly evolving industry, we do need to analyze them rigorously. Less detailed analyses have predicted rapid growth in data center energy use, but without fully considering the historical efficiency progress made by the industry. When we include that missing piece, a different picture of our digital lifestyles emerges.”
对外开放
Opening up
Shehabi 在接受DCD采访时说，可以通过提高行业透明度来进一步改善此类研究，并敦促数据中心运营商联系该论文的研究人员，以机密方式共享其设施的详细信息。行业中的每个人都可能更加开放，但是特别难以获得的数据来自具有内部数据中心的企业以及一般的中国公司。
Speaking to DCD, Shehabi said that such research could be improved further by more transparency in the industry, and urged that data center operators contact the paper's researchers to confidentially share details about their facilities. Everyone in the industry could be more open, but data that was particularly hard to come by was from enterprises with in-house data centers, and from Chinese companies in general.
“多年来，缺乏数据阻碍了我们对全球数据中心能源使用趋势的理解，” Koomey Analytics的合著者Jonathan Koomey说。“这种知识鸿沟使业务和政策规划变得异常困难。”
“Lack of data has hampered our understanding of global data center energy use trends for many years,” coauthor Jonathan Koomey of Koomey Analytics said. 'Such knowledge gaps make business and policy planning incredibly difficult.”
重新校准全球数据中心能源使用估算研究中使用的模型和数据集将公开。
The models and data sets used in the Recalibrating global data center energy-use estimates study will be made public.
研究人员还提出了三项具体的政策建议：
The researchers also made three specific policy recommendations:
通过加强IT能源标准（例如ENERGY STAR），提供财务激励措施以及传播最佳能效做法，来延长当前能效趋势的寿命
Extend the life of current efficiency trends by strengthening IT energy standards such as ENERGY STAR, providing financial incentives, and disseminating best energy efficiency practices
增加对下一代计算，存储和除热技术的研发投资，以减少未来的能源使用，同时鼓励可再生能源的采购，以同时减少碳排放
Increase R&D investments in next-generation computing, storage, and heat removal technologies to mitigate future energy use, while incentivizing renewableenergy procurement to mitigate carbon emissions in parallel
投资于数据收集，建模和监控活动，以消除盲点并做出更可靠的数据中心能源政策决策
Invest in data collection, modeling, and monitoring activities to eliminate blind spots and enable more robust data center energy policy decisions
该文件在结论中警告说：“鉴于数据中心将在未来的能源系统中扮演重要角色，历史上对其能源使用的知识缺乏以及矛盾的发现给决策者提供的混合信号是不可接受的。进入关键的过渡阶段；为确保低碳和节能的未来，我们不能再等十年才能获得下一个可靠的自下而上的估计。”
In its conclusion, the paper warned: 'Given the important role data centers will play in future energy systems, the historical dearth of knowledge on their energy use and the mixed signals given to policymakers by contradictory findings are unacceptable. Global data center energy use is entering a critical transition phase; to ensure a low-carbon and energy-efficient future, we cannot wait another decade for the next reliable bottom-up estimates.'
联系客服
