New Frontiers in Data Center Sustainability

New Frontiers in Data Center Sustainability

Rich Miller writes in DATACENTERFrontier that Beyond Green Power: New Frontiers in Data Center Sustainability can easily be envisioned as these are increasingly populating planet earth.

Above picture is of Large pipes sporting Google’s logo colors move water throughout the cooling plant at the Google’ data center in Douglas County, Georgia. (Photo: Google)

February 3, 2021

New Frontiers in Data Center Sustainability
Get the full report.

Sustainable Construction Strategies

More data center projects will integrate sustainability into design and construction, with early collaboration between teams to minimize the environmental impact of the construction process and create a building with low operational carbon impact, enabling more effective and cost-efficient offset strategies. Design collaboration is essential in seeking to integrate cleaner technologies into the power chain and cooling systems.

Several data center providers are working with CarbonCure, which makes a low-carbon “greener” concrete material for the tile-up walls that frame data centers. Concrete’s durability and strength are ideal for industrial construction, but the production of cement requires the use of massive kilns, which require large amounts of energy, and the actual chemical process emits staggeringly high levels of CO2. CarbonCure takes CO2 produced by large emitters like refineries and chemically mineralizes it during the concrete manufacturing process to make greener and stronger concrete. The process reduces the volume of cement required in the mixing of concrete, while also permanently removing CO2 from the atmosphere.

Waste Stream Accountability and the ‘Circular Economy’

A key priority is tracking the environmental impact of construction components, including a “reverse logistics” process to track the waste stream and disposition of debris. Asset recovery and recycling specialists will become key partners, and the most successful projects will communicate goals and best practices across the contractors and trades participating in each project. The goal is a “circular economy” that reuses and repurposes materials.

Managing packaging for equipment that is shipped to a data center facility is an important and often underlooked facet of waste stream accountability. There are also opportunities in reuse of components and equipment that that can still be productive (although this must be closely managed in a mission-critical environment).

The ability to document a net-zero waste stream impact has the potential to emerge as an additional metric for data center service providers, as customers consider the entirety of their supplier’s sustainability programs.

Green Certifications

As customers ask tougher questions about a providers’ environmental practices and corporate social responsibility policies, certifications may emerge as another avenue for service providers to differentiate themselves.

Several ISO certifications, including ISO 50001 and ISO 14001, which Iron Mountain is certified for across its global data center portfolio, focus on energy management and provide frameworks that can assure stakeholders that the provider is considering energy impact and environmental goals in audits, communications, labeling and equipment life cycle analysis.

Water Conservation and Management

Amid changing weather patterns, many areas of the world are facing drought conditions and water is becoming a scarcer and more valuable resource. Data center operators are stepping up their efforts to reduce their reliance on potable water supplies.

Sustainable water strategies include both sourcing and design. On the sourcing front, several Google facilities include water treatment plants that allow it to cool its servers using local bodies of water or waste water from municipal water systems. Data center districts in Ashburn (Va.), Quincy (Washington) and San Antonio offer “grey water” feeds that provide recycled waste water to industrial customers.

On the design front, more providers are choosing cooling systems with minimal need for water, while others are incorporating rainwater recovery strategies that capture rain from huge roofs or parking lots and store it on site, reducing potential burden on local water systems.

Matching Workloads to Renewable Energy

Google has been a leader in the use of artificial intelligence and sophisticated energy provisioning to match its operations to carbon-free energy sources. The company recently said it will power its entire global information empire entirely with carbon-free energy by 2030, matching every hour of its data center operations to carbon-free energy sources. This marks an ambitious step forward in using technology to create exceptional sustainability.

Google can currently account for all its operations with energy purchases. But the intermittent nature of renewable energy creates challenges in matching green power to IT operations around the clock. Solar power is only available during daylight hours. Wind energy can be used at night, but not when the wind dies down. Google created a “carbon-intelligent computing platform” that optimizes for green energy by rescheduling workloads that are not time-sensitive, matching workloads to solar power during the day, and wind energy in the evening, for example. The company also hopes to move workloads between data centers to boost its use of renewables, a strategy that offers even greater potential gains by shifting data center capacity to locations where green energy is more plentiful, routing around utilities that are slow to adopt renewables.

Google has pledged to share its advances with the broader data center industry, providing others with the tools to reduce carbon impact. Continued instrumentation of older data centers is a key step in this direction.

Eliminating Diesel Generators

New Frontiers in Data Center Sustainability
A backup generator at a Microsoft data center in Virginia. (Photo: Rich Miller)

Microsoft recently announced plans to eliminate its reliance on diesel fuel by the year 2030, which has major implications for the company’s data centers, many of which use diesel-powered generators for emergency backup power. With its new deadline, Microsoft sets in motion a push to either replace its generators with cleaner technologies, or perhaps eliminate them altogether by managing resiliency through software.

Eliminating expensive generators and UPS systems has been a goal for some hyperscale providers. Facebook chose Lulea, Sweden for a data center because the robust local power grid allowed it to operate with fewer generators. In the U.S., providers have experimented with “data stations” that operate with no generators on highly-reliable locations on the power grid.

There are four primary options companies have pursued as alternatives to generators — fuel cells, lithium-ion batteries, shifting capacity to smaller edge data centers that can more easily run on batteries, and shifting to cloud-based resiliency.

Fuel Cells and On-Site Power

Microsoft has successfully tested the use of hydrogen fuel cells to power its data center servers. The company called the test “a worldwide first that could jump-start a long-forecast clean energy economy built around the most abundant element in the universe.”

Microsoft said it recently ran a row of 10 racks of Microsoft Azure cloud servers for 48 hours using a 250-kilowatt hydrogen-powered fuel cell system at a facility near Salt Lake City, Utah. Since most data center power outages last less than 48 hours, the test offered a strong case that fuel cells could be used in place of diesel generators to keep a data center operating through a utility outage.

Some companies, like Equinix and eBay, have deployed Bloom Energy fuel cells to improve reliability and cut energy costs, but have powered them with natural gas. The use of biofuels looms as another potential avenue to pair fuel cells with renewable sourcing.

Energy Storage

New Frontiers in Data Center Sustainability
An illustration of the Tesla Megapack, which provides 3 megawatts of energy storage capacity. (Image: Tesla)

Utility-scale energy storage has long been the missing link in the data center industry’s effort to power the cloud with renewable energy. Energy storage could overcome the intermittent generation patterns of leading renewable sources. Solar panels only generate power when the sun is shining, and wind turbines are idle in calm weather. Energy storage could address that gap, allowing renewable power to be stored for use overnight and on windless days.

A new project in Nevada will showcase a potential solution from Tesla, the electric car company led by tech visionary Elon Musk. Data center technology company Switch will use new large-scale energy storage technology from Tesla to boost its use of solar energy for its massive data center campuses in Las Vegas and Reno. It is a promising project in pioneering a holistic integration of renewable power, energy storage and Internet-scale data centers.

Talking Sustainability With Experts

Don’t miss the last installment of this series that features a conversation on the future of sustainable data centers. Data Center Frontier Editor Rich Miller discusses the topic with Kevin Hagen, Director, Corporate Responsibility at Iron Mountain, and Alex Sharp, Global Head of Design & Construction — Data Centers at Iron Mountain.

It’s a preview of the upcoming webinar  where these experts will discuss sustainability strategies for greener data centers.

And catch up on the first entry here, followed by an exploration of the power of the “negawatt.” 

Download the full report, Green Data Centers and The Sustainability Imperative, courtesy of Iron Mountain, to explore how climate change and a greening of data centers is changing the industry.

If data are new gold, governance can safeguard society

If data are new gold, governance can safeguard society

China Daily Global in an article titled ‘If data are new gold, governance can safeguard society’, perhaps domestically, but says it all about what to expect in the future relationship of China with say countries of the MENA region.

If data are new gold, governance can safeguard society

By Liu Xiaochun | China Daily Global | Updated: 2021-01-18

The Central Economic Work Conference held in December outlined certain key tasks in eight major aspects for this year. These include strengthening efforts in antitrust and preventing the disorderly expansion of capital.

It was clearly pointed out in the meeting statement that the collection, usage and management of data shall be improved.

With robust growth of the “new infrastructure” sector, particularly the application of 5G and the internet of things, digital technology will find applications in all walks of society and will bring significant change to people’s way of living.

While appreciating the positive effect that digital society may bring, it is important to fully acknowledge and evaluate the risks that interconnectivity of data may bring and pay attention to data governance.

As digital technology is highly penetrative and spreads widely, the risk of digital technology can be widely disruptive and can go beyond personal privacy. It thus requires precautionary regulatory measures to manage or pre-empt such risks.

There are key issues and risks in data connectivity, and it is important to strike a proper balance between breaking the information silo and data security.

On the one hand, it is important to clarify which part of the society will guide the connectivity of data, be it the government, technology firms or other institutions. For example, the building of smart cities will require data collection from a great number of sectors and departments. It is crucial to make clear who will be responsible for collecting and managing them.

On the other hand, how data can be categorized and managed is another emerging issue. In governing smart cities, new data of all kinds emerge every second. The idea of smart city construction, building industrial internet and digital China cannot be realized without data from all departments and organizations going online.

Yet, with all these key data openly accessible online, inadequate or improper management of these data may pose a possible threat to public security, the police, or even to social and national security.

Both governance and the internet of things across all industries should take the management of public data into account. At the same time, the arithmetic model, a key technology in artificial intelligence, may amplify potential risks in information spreading with no targeted audiences.

There is also the risk of giant internet and technology companies adopting a winner-takes-all approach in data collection. Conventional monopoly usually means taking monopoly of one particular type of products or at most, a certain industry. The new winner-takes-all approach would mean exclusive owning of all data on one particular platform by a certain enterprise.

Online platforms in fields such as e-commerce, digital payments, and delivery services may even gain access to huge amount of social data in the name of innovation or breaking up information silo. Such data may be related to personal, business or even government information.

Should such platforms or online behemoths land in major trouble, or face some unforeseen risks, massive systemic disruptions could unsettle or destabilize society. And with the growth of 5G, the number of such businesses is expected to grow.

A number of steps will likely be taken to strengthen data governance. Control of data risks should be raised as part of State governance efforts. Any arbitrary collection of personal information and data should be prohibited.

The issue of data categorization needs to be resolved through legislative efforts in this field. A number of suggestions have been made in legislation regarding personal information protection, which is very necessary.

Categorization should be made for data under digital economy.

First, special attention should be given to managing data regarding public security, finance and people’s livelihood, and how they can be made accessible on internet platforms and how such data can be used.

Second, the responsibility of data management should be specified, and ownership and usage rights to data clarified.

Third, legal liability in data use and transaction must be made clear.

Fourth, as data management is a new and emerging sector yet closely related to national security, social stability and a steady running of economic activities, a special regulatory department or mechanism should be set up with powers of oversight.

At the same time, a category-specific, more proper oversight on artificial intelligence is also needed, particularly a more targeted regulatory model for algorithms developed by various businesses.

An overhaul of personal data already collected once all the aforementioned systems are in place would be in order.

Mechanism for the oversight and management of super-giant data platforms should be set up. On the one hand, objective views are needed about the monopolies taken by super-giant digital platforms.

These platforms also bear public service functions, differentiating them from industrial or commercial monopolies. Concentration of platforms may also help add on commercial competitiveness and social efficiency.

Take third-party payments as an example. To ensure unimpeded payments, various market participants tend to gather on one payment platform. If communications across different telecom companies cannot be realized, only one telecom platform will eventually survive.

Such logic also applies to third-party transactions, which explains why even though the regulators concerned issued a number of licenses, only a few survived. And there are reasons behind why only those few did manage to survive.

First, the survivors are those that are supported by the banks’ unified payment services. Second, the companies specialized in integrated payment services has become a solution for third-party payment platforms banning one another.

Super-giant platforms will likely continue to increase as digital society grows. Concentration of multiple services in a single platform may make business sense for market share-minded companies. But it is debatable if this is the right path to digital transformation of society.

So, proper regulatory measures and oversights are needed in helping such platforms to grow with society in a responsible manner. This is why, oversight mechanisms are needed, as platform enterprises can’t achieve this on their own through self-regulation.

Meanwhile, all data collected by platform businesses are related to society’s various publics and therefore should not be treated as commercial assets.

The article is a translation of a comment from the Bund Summit by Liu Xiaochun, the deputy dean of the Shanghai Finance Institute.