GreenBay Digest

Huawei’s iCooling@AI solution is dedicated to improving the power usage effectiveness (PUE) of data centers, resulting in an energy reduction of 8 to 15 percent. This substantial saving contributes to creating a greener and smarter future.

Data centers play a crucial role in supporting the vast array of smart services accessed by billions of Internet users worldwide. With the rapid advancement of technologies like 5G, IoT, and cloud computing, the demand for data has skyrocketed, leading to the growth of the data center industry.

Energy Efficiency Challenges
Data centers consume a significant amount of electricity, accounting for approximately 1 percent of global energy expenditure. However, despite the exponential growth in data center computing instances, power consumption has only increased by 6 percent since 2010. This trend indicates a positive improvement in energy efficiency, thanks to collaborative efforts by governments and enterprises.

iCooling@AI: Enhancing Energy Efficiency
To address the challenges of energy consumption in data centers, Huawei has developed iCooling@AI, an innovative solution that optimizes power efficiency through the integration of big data and AI. By leveraging intelligent algorithms and continuous self-optimization, iCooling@AI enables data centers to reduce their PUE by 8 to 15 percent. This intelligent control system replaces manual adjustments and dynamically adjusts cooling policies based on real-time data analysis.

Partnership Success: China Unicom Henan
China Unicom Henan, a leading operator, has implemented Huawei’s iCooling@AI solution in their data centers. By utilizing iCooling@AI’s capabilities, China Unicom Henan has achieved significant improvements in power efficiency, resulting in enhanced cost-effectiveness and reduced environmental impact.

The financial industry is undergoing significant disruptions driven by emerging technologies like 5G, cloud computing, and big data. To align with China’s “dual-carbon” goal and reduce environmental impact, the Shanghai Stock Exchange (SSE) built the Jin Qiao Data Center, Asia’s most energy-efficient data center in the financial sector.

Efficient cooling is crucial for a green data center. The Jin Qiao data center, serving as the SSE’s main operations center, houses 186,000 racks and consumes over 120 million kWh of electricity annually. To improve energy efficiency, the SSE selected Huawei’s iCooling@AI solution, which uses AI algorithms to optimize cooling operations based on real-time data, climate conditions, and IT loads. This smart cooling technology reduces the power usage effectiveness (PUE) by 13% annually, contributing to significant energy savings.

iCooling@AI enables visualization of cooling links, equipment connections, and transforms manual adjustments into intelligent collaboration. By integrating iCooling@AI into the Jin Qiao Data Center, Huawei and SSE have successfully decreased the annual average PUE by 13%. This achievement was recognized with the National Cloud Computing Center Science and Technology Award.

The Jin Qiao Data Center is Huawei’s first endeavor to integrate iCooling@AI into a hyper-scale data center in the financial sector. Moving forward, Huawei aims to collaborate with the Shanghai Stock Exchange to build a green and low-carbon data center model for the financial industry, contributing to sustainable and green finance.

Nigeria, the most populous country in Africa, has experienced steady economic growth over the past two decades and currently boasts the largest economy on the continent. With a staggering 90 million internet users, Nigeria has the highest number of African internet users and is ranked tenth globally. The Information and Communications Technology (ICT) industry is a major contributor to Nigeria’s strong economic growth and has facilitated the rapid development of various industries such as e-commerce, online banking, and digital government.

The burgeoning digital economy has also given rise to the development of data center infrastructure in the country. According to Xalam research, Internet Data Center (IDC) capacity in Sub-Saharan Africa (SSA) was recorded at 230 MW in 2020, with an average annual growth rate of 30% over five consecutive years. Capacity is projected to increase to anywhere between 350–450 MW by 2024. This remarkable potential has attracted enterprises to invest in IDCs and build large-scale data centers in Nigeria.

CloudExchange, a leading system integrator in West Africa, recognized the tremendous market potential and decided to enter the IDC market. As a new entrant, CloudExchange needed to differentiate itself and establish a foothold in a fiercely competitive environment. To stand out, the company partnered with Huawei to offer superior IDC services with faster response times to requirements, more flexible capacity expansion, and more reliable data centers overall.

To speed up the construction process, CloudExchange chose a Huawei prefabricated modular data center solution. With most devices pre-installed and pre-tested in the factory, the data center only needs to be hoisted onsite, reducing the construction period of the entire project to just seven months. The Huawei FusionDC solution also supports online capacity expansion, with each layer functioning as an independent data center unit, allowing for quick and interruption-free capacity expansion.

CloudExchange’s new data center is the first Uptime Institute Tier IV prefabricated data center in Africa, which means it offers ultra-reliable uptime. Huawei worked closely with CloudExchange to deploy dual-socket diesel generators, dual-fuel tanks, and equipment rooms in a limited space, maximizing space utilization and gaining Uptime Institute Tier IV certification in a significantly streamlined process.

Glad Dibetso, CEO of CloudExchange, expressed his satisfaction with the fruitful partnership with Huawei, saying that their cooperation has quickly obtained market recognition and certification from the Uptime Institute. He hopes that their project will continue to boost ICT industry development in West Africa and propel the ongoing rise of the entire continent.

The Gui’an Huawei Cloud Data Center is the world’s largest Huawei Cloud data center with an operational capacity of over 1 million servers. It serves as a key facility for Huawei Cloud, Huawei BP&IT and consumer cloud services. Located in Gui’an New District, Guizhou Province, construction began in August 2017 and it is one of two main Huawei Cloud data centers in China.

The phase 1 construction area covers approximately 480,000 square meters and is divided into three subareas. All 51 buildings in phase 1 have been completed with nine being used as the data center facility and the rest as auxiliary facilities.

Huawei Cloud’s Gui’an data center is designed with green and intelligent technologies to reduce energy consumption and GHG emissions. With a PUE of 1.12, it is one of the most efficient in the industry.

Some of the measures taken include:

• using direct ventilation for natural cooling,
• developing a medium-temperature chilled water system that reuses waste heat,
• introducing liquid cooling technology to increase efficiency and reduce PUE,
• applying AI to balance loads among servers and increase resource utilization, and
• replacing copper components with power semiconductors to reduce power loss.

When operating at full capacity, the data center is expected to save 1.01 billion kWh of electricity per year and offset more than 810,000 tons of carbon emissions.

The Uptime Institute’s Global Data Center Survey 2022 suggests that over the next five years, data centers should play a role in energy conservation and carbon reduction, rather than being a hindrance to it.

With this in mind, Huawei plans to change the way people used to plan, build and operate data centers during the construction. Huawei’s Dongguan Cloud Data Center T1 project makes noticeable progress in energy conservation and carbon reduction. Put into operation in 2019, the project’s PUE reached 1.28 in 2021.

In Part 1 of this article, we have seen how digital technology enablement contributed to the success of the T1 project. Today, we look at other advanced energy-savings technologies which were employed.

AI-based PUE optimization
The T1 project employs an AI-based iCooling solution to decrease the energy consumption of the cooling system, which typically uses over 30% of a data center’s energy. This solution constantly monitors the energy consumption of all cooling subsystems and delivers the most efficient cooling strategy based on factors like current IT load and outdoor temperature.

Green construction and material recycling
The construction of traditional data centers often results in a high amount of waste and excessive water and power consumption. To address these issues, the T1 project employs a prefabricated modular solution where core components and systems are manufactured and pre-tested in a factory before being delivered and assembled on-site like LEGOs. This approach significantly reduces water usage and construction waste by 80%, reduces carbon emissions by 90%, and recycles up to 85% of materials, promoting eco-friendly construction and maximizing resource recovery.

Efficient and energy-saving devices
As data centers consume a large amount of energy, the use of energy-efficient and energy-saving devices is crucial to reducing energy consumption. The Huawei FusionPower6000 is utilized for power supply and distribution, providing a single source of power for each row, which boosts end-to-end efficiency to 95.5%. The cooling system is equipped with high-performance fans and wet film humidifiers to enhance energy efficiency. Additionally, smart modules are arranged in the equipment room to divide cold and hot aisles, optimizing heat exchange efficiency.

By using energy-saving technology, data centers can reduce their impact on the environment, contribute to a more sustainable future, and demonstrate their commitment to environmental responsibility.

Source: Huawei Digital Power

Huawei’s Dongguan Cloud Data Center T1 project (T1 project for short) is among the two green data centers in China to have been selected as a typical application case of key energy-saving technologies 2021 by the National Energy Conservation Center for the large amounts of energy it has saved during operation. Put into operation in 2019, the project’s PUE reached 1.28 in 2021.

Today, we look at how digital technology enablement contributed to the success of the T1 project.

Due to insufficient simulation of air ducts and airflows at the design phase and backward onsite assembly and construction, traditional data centers have a higher PUE than the designed one. By contrast, the T1 project uses digital design, simulation, and manufacturing, reaching a designed PUE in actual operation.

Next week, we will look at other advanced energy-savings technology which was employed in the T1 project.

Source: Huawei Digital Power

Data centers consume a great deal of energy and resources, and they are one place that many organizations seek to cut back – with the goal of eventually reaching net-zero carbon emissions. Globally, leading data center operators have committed to carbon neutrality and science-based targets for emissions reduction by 2030. The challenge is how to achieve this. Huawei has outlined four core trends for the future direction of data centers.

Solving the challenge of growing demand for data with low-carbon solutions, Huawei’s next-gen simplified, green, intelligent, and secure data center infrastructure solution is only one of the solutions. Reducing carbon emissions is essential for the industry and an area in which innovation focuses on converging software and hardware as well as cross-product-domain collaboration. With so many technical innovations, expertise and best practice at our disposal, the journey to zero-carbon data center is now achievable.

After-sales maintenance helps ensure that a customer gets as much use and value as possible out of their purchase. A great product should be backed up with excellent maintenance support as this provides a positive experience to customers.

The business reasoning behind after-sales support is that it can contribute to brand loyalty and positive word-of-mouth. An example of a brand which understands this reasoning is Huawei.

Huawei Hi-Care services provide a variety of maintenance services for customers. Huawei Hi-Care provides industry-leading Technical Support Centers (TAC), global service organizations, hourly spare parts supply capabilities, standard Issue to Resolution (ITR) processes, and advanced IT tools to respond to customer service requests in a timely and efficient manner.

Huawei has always been committed to providing customers with more professional and intelligent maintenance services. Huawei Hi-Care services go beyond traditional phone or email support as customers can seek help from multiple channels such as the intelligent Q&A chatbot and the technical support community. In addition, customers can use the intelligent maintenance center ServiceCare to manage inventory and service requests, and quickly obtain product technical information and service reports.

At GreenBay, we believe that an effective maintenance strategy is concerned with maximising equipment uptime and facility performance while balancing the associated resources expended and ultimately the cost. When a brand offers maintenance services (such as Huawei Hi-Care) which can assist in getting sufficient return on your asset investment, it is an indication that your maintenance strategy is on the correct trajectory to be successful and fuss-free.

A data center’s power usage effectiveness or PUE, is seen as a key indicator for evaluating its energy efficiency. The PUE value rises as the data center becomes less efficient. Cooling, a key component of a data center, is one of the main factors which can address the issue of energy efficiency.

Cooling is closely related to equipment heat dissipation, equipment configuration, facility environment, and external climate conditions. After Operations and Management (O&M) reaches a certain maturity level, hardware energy saving or human experience-based simple optimization cannot further reduce power consumption.

Therefore, Huawei has leveraged on machine learning to develop its iCooling intelligent thermal management solution for data centers. The iCooling solution incorporates deep learning to analyse historical data and identify key factors which affect energy consumption and create a PUE prediction. An optimisation algorithm then establishes the ideal parameters which are transmitted to various control systems.

The key technologies used in the iCooling solution include:

• Big data collection
  Given the complexity of data center cooling systems, information about the power supply system, cooling system, and environment parameters must be collected.
• Data governance and feature engineering
  First, a mathematical tool is used to perform data governance on the raw data collected, providing high-quality data for subsequent model training. Second, feature engineering is performed on large amounts of raw data to identify the key parameters that affect PUE.

Creating a PUE model using a neural network:
Neural networks are a set of machine learning algorithms that can simulate the cognitive behavior of interactions between neurons. Deep neural networks can play a role in increasing the cooling efficiency of data centers. The machine learning algorithms of these networks can find the relationships between parameters of different pieces of equipment and systems. A mathematical model or the PUE model of the data center is created based on large amounts of data from sensors.

Inference and decision-making using genetic algorithms:
Based on the input PUE model and the operating data collected in real time, the algorithms find the best policy in four steps: parameter traversal and combination, service rule assurance, calculating the energy consumption of the cooling system, and selecting the optimal policy.

At Huawei’s cloud data center Langfang in China, the deployment of iCooling has resulted in a PUE that is eight percent lower than it previously was, which translates to major savings in annual power costs. At a China Mobile data center in Ningxia, the introduction of iCooling technology has reduced the data center’s total energy consumption by 3.2%, saving more than 400,000 kWh of electricity a year. According to Huawei, as data centre loads increase and AI learning capability improves, six million kWh of electricity will be saved in the data center every year, the equivalent to a reduction of about three million kilogrammes of carbon dioxide emissions.

With extensive experience and technical expertise in global data center construction, Huawei has proven that the power of AI and big data is the key to create greener data centers.

Source: Huawei

Protective relays are decision-making elements in the protection scheme for electrical power systems. They monitor circuit conditions and initiate protective action when an undesired condition is detected.

Protective relaying serves many functions including isolating faulted circuits or equipment from the remainder of the system so the system can continue to function, limiting damage to faulted equipment, minimizing the possibility of fire or damage to adjacent equipment, and minimizing hazards to personnel.

According to Reg. 110(4), ER (Electricity Regulations) 1994; any protective relay and device of an installation will need to be checked, tested and calibrated by a competent person at least once every two years, or at any time as directed by the Energy Commission.

Why is protective relay testing important?
When they are required to operate because of a faulted or undesirable condition, it is imperative that protective relays function correctly. A strong maintenance and test program will keep your protective relays in a high state of readiness ensuring they respond properly to normal and abnormal conditions and help you avoid equipment damage, prolonged downtime and employee injuries.

When protection relay testing is not carried out on a regular basis, the risk of dangerous electrical faults increases, potentially resulting in damage to equipment and harm to personnel.

Types of testing
There are three main types of protective relay testing (apart from testing after a fault occurs):

• Bench testing – This involves investigating the relay on its own and that it matches the design. This prevents more expensive and more time-consuming problems from occurring at later stages in a project.
• Commissioning testing – Once the electrical system has been constructed, commissioning the relay involves making sure the larger system functions as expected. For example, when the protection relay is attached to the switchgear, it should function as expected, react to interlocks and other simulated conditions. After this, the relay’s function will have been demonstrated.
• Maintenance testing – When maintenance testing is carried out, complete design intention is assumed, but relay behaviour must be checked under operation. Apart from relay-specific failures, the relay cannot detect changes in system characteristics, such as downstream network loads being changed over time. Such long-term changes may require the relay to be reprogrammed to ensure expected operation is maintained.

Safety is everything
Electrical safety is not an issue to be taken lightly. According to statistics released by the Fire and Rescue Department, a total of RM3.31 billion worth of properties were destroyed in fire-related incidents in 2018. Of that number, faulty electrical sources were at the very top of the list with over 5,300 cases recorded over the last three years.

This is the reason why at GreenBay, we take a proactive stance when it comes to electrical safety:

• All electrical works are carried out by a team of competent and qualified engineers.
• The electrical works shall be carried out in accordance with rules and safety regulations and best working practices.
• All installation shall be maintained in good working order and safety precautions shall be observed at all times to prevent danger.