更标准化 | 使用新技术的下一代数据中心

关注标记「DeepKnowledge」设为星标

更多干货等你发现!



使用新技术的下一代数据中心

NEXT GENERATION DATA CENTRES FOR NEW GENERATION TECHNOLOGY

 

BY NEXT DC  | May 05, 2018 


作为澳大利亚领先的独立数据中心运营商,NEXTDC拥有Uptime Institute III级认证设施。NEXTDC已经加大了对下一代数据中心的投入,随之而来的是实现了B2布里斯班和M2墨尔本Uptime Institute IV级认证。

Australia’s leading independent data centre operator with a nationwide network of Uptime Institute Tier III certified facilities, NEXTDC have upped the ante with our next generation data centres, achieving Uptime Institute Tier IV certification for B2 Brisbane and M2 Melbourne.



我们的客户对连续正常运行时间、灵活的连通性和可伸缩性的需求不断增加,因此在规划下一代数据中心时,我们要特别关注关键基础设施的卓越弹性。 

When planning the next generation of our data centres, we saw an opportunity to improve the already exceptional resilience of the critical infrastructure we deliver our customers who have increasing demands for continuous uptime, flexible connectivity and scalability.


 

到达TierIV的旅程
The journey to Tier IV


我们新的运营可持续性和容错数据中心Uptime Institute(Tier IV)设计认证表明,持续创新已经嵌入我们的DNA里。  这是对我们的设施能够承受单个设备故障或配电路径中断和维护IT操作能力的宝贵认可。

Our new Uptime Institute certifications in operational sustainability and fault tolerant data centre design (Tier IV) demonstrates that continuous innovation is embedded in our DNA. A valuable endorsement of our facilities’ ability to withstand individual equipment failures or distribution path interruptions and maintain IT operations.


Tier IV认证目前是Uptime Institute机构的最高级别认证。它用于一致性地评估单个数据中心的性能和在构建Tier III标准的并发可维护的标准之上。

Tier IV certification is currently the peak of the tier classification system created by Uptime Institute to consistently evaluate individual data centre performance and builds on the Tier III standard for concurrent maintainability.


Tier IV认证的主要区别是容错。多个独立、物理隔离的系统提供冗余容量和不同的分布路径可以同时支撑关键的运行环境。 Tier IV还提供了连续冷却,以及所有N+1路径和关键设备之间的防火隔离和泄漏遏制。

The main feature of Tier IV certification is Fault Tolerance, where multiple independent, physically isolated systems provide redundant capacity and diverse distribution paths to simultaneously serve the critical environment. Tier IV also provides continuous cooling, as well as fire segregation and leak containment between all N+1 paths and critical equipment.


做到这一切会有非凡的弹性。由意外或计划中的事件引起的单个设备故障或配电路径中断不会影响IT业务的运行。

Achieving all this results in exceptional resilience. Individual equipment failures or distribution path interruptions caused by unexpected or planned events, will not impact IT operations.


从我们开始规划第二代数据中心的那一刻起,我们就为自己设定了一个积极但重要的目标。我们坚信,我们将以比第三级更低的每兆瓦成本建造我们的第二代四级设施。 作为一个例子,对于B2,我们设计并安装了建筑外的水管,这释放了多排额外的机架。此外,通过在建筑外安装管道,我们默认实现了灾难性的泄漏容忍度。

From the moment we began planning for our second generation of data centres we set ourselves an aggressive, but important goal. It was our firm belief that we would build our second generation of Tier IV facilities at a lower cost per MW than Tier III. As a simple example, for B2 we designed   and later installed the water pipes outside the building, which freed up multiple rows of additional racks. In addition, by installing the pipes outside the building we achieved catastrophic leak tolerance by default.


“Tier IV的原则是保持设计美观简洁,并保持多路径分离,使其有容错能力”。NEXT DC的工程主管杰夫·范·泽顿说“为什么我们做这个?” 。“我们建设和运维的数据中心是新社会的基础设施,它们必须满足最高的可靠性标准——最重要的是,创新是我们NEXTDC的核心主旨之一。” 

“Tier IV is about keeping the design elegantly simple, and keeping the paths separated to make it fault tolerant.” “Why do we do this?” says NEXTDC’s head of engineering Jeff Van Zetten.  “The data centres we build and operate are the infrastructure for the new society, they have to meet the highest reliability standards – on top of that, innovation is part of our core charter at NEXTDC.”


第二代数据中心项目代表了NEXTDC的设计从目前的Tier III向更强大的Tier IV演变。 我们为所有下一代设施指定了标准实践。其中三分之二的设施已经获得了Tier IV的认证。 这些设施将在整个生命周期内得到黄金标准运营组织的支持。悉尼地区的设施一旦建好,也将效仿。 

The second-generation data centre projects represent the evolution of NEXTDC’s data centre design from the current Tier III standard to the even more robust Tier IV. We have a standard practise in place for all our next generation facilities. Two out of three have already received Uptime Institute Tier IV Constructed Facility, which will be supported throughout their life by a Gold standard operational organisation. It is intended that Sydney, once completed, will follow suit.


“我们的第一代数据中心已经非常接近Tier IV”,我们的工程主管杰夫·范·泽顿说到。 “我们已经有了Piller动态UPS系统支持整个冷却系统,它们可以持续冷却。我们已经将发电机房和UPS机房分开”。

“Our generation one data centres are incredibly close to Tier IV already,” commented our Head of Engineering and Design, Jeff Van Zetten. “As we already have the entire cooling system supported by the Piller rotary UPS system, they are continuously cooled. We had already started fire segregating the generators and UPS rooms.”



“我们建设和运营的数据中心是新社会的基础设施。他们必须满足最高的可靠性标准——除此之外,创新是我们NEXT DC 的宗旨之一。” 

“The data centres we build and operate are the infrastructure for the new society. They have to meet the highest reliability standards-on top of that, innovation is part of our core charter at NEXT DC”


解决方法
The solution 


为了成功实现Tier IV的容错要求,NEXT DC工程师团队从两个不同的角度去解决:在电气系统设计中增加差动故障和隔离变压器;以及每根电源线的完全防火隔离。

In order to successfully achieve the fault tolerance requirement that a Tier IV Uptime Institute certified data centre demands,   the NEXTDC engineering team approached it from two different angles: The addition of differential fault protection and isolation transformers added to the design of the electrical system; and the complete fire-segregation of each power string.


许多Tier IV的设计依赖于显著增加冗余的基础设施来达到容错的目的。而我们第二代设施只要一个额外的RUPS单元(+1)添加到支持整个电气系统的单元集群里就行。因此,如果一个电气故障在系统内的任何一个地方发生,这部分很快就会孤立——包括火灾——这意味着尽管系统支持的负载数量已经从N+1 减到N,但是减后的剩余系统仍不受影响。系统的其余部分(剩余的负载容量为N)会自动增加,保证连续供电。 

Where many Tier IV designs rely on a significant increase in redundant infrastructure to achieve Tier IV status for fault tolerance, our second-generation facilities only required an additional RUPS unit (+1) to be added to the total number of units required to support the electrical system in each facility (N). As a result, if an electrical fault develops on any one part of the system, that part is very quickly isolated — and in the case of fire, contained — meaning that although the amount of systems supporting the load has been reduced from N+1 to N, the rest of the system remains unaffected. That ‘rest of the system’ (i.e. the remaining N capacity) automatically ramps up, ensuring continuous availability of power throughout the data centre without interruption.


在实现系统容错时要考虑的关键因素之一是冷却。随着冷却技术的最新进展,NEXTDC的工程团队开发了一种简化且高度隔离的冷却系统解决方案,该方案具有容错能力,满足每个数据中心的冷却需求。一个共同的特点是使用一系列非常高效的独立模块冷却单元。 

One of the crucial factors to consider in achieving the system’s fault tolerance is cooling. With recent advances in cooling technology, NEXTDC’s engineering team developed a simplified and highly segregated cooling system solution, which are fault tolerant and meet each data centre’s cooling requirements. One common feature is the use of a series of incredibly efficient, stand-alone modular cooling units.


“通过研究世界各地的Tier iV等级设施,我们学到的一件事是,你不需要花更多的钱。以B2为例,我们在建筑外安装和设计了水管,腾出了多排额外的机架。通过将管道置于结构外部,我们实现了默认的灾难性泄漏容忍。”

“One of the things we have learnt by studying Tier IV around the world is that it doesn’t have to cost you more. As a simple example, for B2 we installed and designed the water pipes outside the building, freeing up multiple rows of additional racks. By putting the pipes outside the structure, we achieved catastrophic leak tolerance by default.”



 结论

The outcome  


构建一个包含N+N或2N+1设计的Tier IV数据中心的成本,通常只对银行等依赖数据的大型组织具有经济可行性。由于NEXTDC致力于提供最具竞争力的价值,因此并不需要大量的额外投资来增加潜在的可靠性。

The cost of constructing a Tier IV data centre incorporating an N+N or 2N+1 design, was generally only economically viable for large data-dependent organisations such as banks. As NEXTDC are committed to delivering the best competitive value, the potential increase in reliability did not justify the substantial additional investment required.


然而,在我们工程师团队结合最前沿的技术创新下,我们几乎以Tier III的建设成本建立了Tier IV的第二代设施。

However, with the combination of cutting-edge technologies with the innovative engineering and design work carried out by the NEXTDC engineering team, we built our second- generation facilities to Tier IV fault tolerance standards at almost the same cost as a Tier III facility.



这意味着,对于任何澳大利亚企业来说,通过最大化正常运行时间和降低因中断造成的财务开支,以最大限度地减少我们对数据的依赖所带来的风险,拥有超常的可靠性和弹性现在是一个经济上可行的选择。

The result of this means having exceptional levels of reliability and resilience is now an economically viable option for ANY Australian business looking to minimise the risk associated with our reliance on data by maximising uptime and lower financial overheads caused by outages.


DeepKnowledge


翻译:

邓秋实

中科大国家同步辐射实验室

DKV(Deep Knowledge Volunteer)创始成员


公众号声明:

本文并非官方认可的中文版本,翻译学习参考,不用于任何商业用途,版权归原文及作者所有,文章内容请以英文原版为准



公众号 | 视频号 | 头条号| 西瓜视频 |BiliBli

首页    更标准化 | 使用新技术的下一代数据中心
设计
管理
运维
设备
电气
暖通
控制
碳中和
储能

新闻动态