Global IT giants are using AI to help data centers save energy and reduce emissions.

Data centers power the applications, websites and services used by billions of people around the world every day, and they can be dangerous places for the workers who build and maintain them. Workers must sometimes service electrical equipment in the data center while its power is on. They may be exposed to chemicals such as chlorine, which is used as a disinfectant in water circulated through the liquid cooling systems of computers and servers. In June 2015, five people had to be taken to the hospital after a chlorine leak occurred at an Apple data center in Maiden, North Carolina.

Now, data centers are more secure than ever. But in search of forward-looking solutions, some tech giants say they are exploring how to apply artificial intelligence to prevent security issues. For example, Microsoft is developing an artificial intelligence system that can analyze data from various sources and generate alerts for data center construction and operations teams to "prevent or mitigate the impact of security incidents." A complementary but related system is also being developed in an attempt to detect and predict impacts on data center construction plans.

"These initiatives are in early testing and are expected to begin scaling to our production environments later this year," a Microsoft spokesperson told the outlet via email.

Meta also claims to be studying how artificial intelligence can predict how its data centers will operate under "extreme environmental conditions" that could lead to unsafe working conditions. The company said it has been developing physical models to simulate extreme conditions and feeding that data into AI models responsible for optimizing server power consumption, cooling and airflow.

Meta spokesperson told the media: "Our data centers hold a large amount of operational data, and in some areas there is a high frequency of sensors built into servers, racks and data halls." "Every server and network device Taking on different workloads will consume different power, generate different heat, and create different amounts of airflow in the data center. Our [infrastructure] team collects all the data from each server and then develops an AI model that can Allocate our servers and racks in data centers and send workloads to those servers to optimize [for] performance and efficiency."

Of course, besides security, businesses have other motivations to secure data The center is kept in top condition. Power outages are expensive—and increasingly frequent. One-third of data center owners and operators admitted to experiencing a major outage in the past 12 months, according to a 2020 survey by IT consulting firm IT Uptime Institute. One in six people claimed their outages cost more than $1 million, up from one in 10 in 2019.

Meta, which operates more than 20 data centers around the world, including new projects in Texas and Missouri, said it will build 50 to 100 new data centers per year for the foreseeable future.

AI also promises cost savings by finding energy-saving opportunities in data centers that are often unknown, another attractive aspect for businesses. In 2018, Google claimed that artificial intelligence systems developed by its DeepMind affiliate were able to save an average of 30% energy compared to the historical energy use of its data centers.

When contacted for comment, DeepMind said it had no updates to share beyond the initial announcement. IBM and Amazon did not respond to inquiries. But both Meta and Microsoft say they are now using AI to make similar power adjustments.

Microsoft launched an artificial intelligence "anomaly detection approach" in late 2021 that uses telemetry data from electrical and mechanical equipment to measure and mitigate abnormal power and water events within data centers. The company also uses AI-based methods to identify and fix issues with data center electrical meters and determine ideal locations to place servers to minimize wasted power, network and cooling capacity.

Meta says it has been using reinforcement learning to reduce the amount of air pumped into data centers for cooling purposes. At a high level, reinforcement learning is an artificial intelligence system that learns to solve problems through trial and error. Most data centers use outdoor air and evaporative cooling systems, so optimizing airflow is a top priority.

Reducing the environmental footprint is an added benefit of the energy regulating AI system. According to a report by the Environmental Investigation Agency, data centers consumed approximately 1% of global electricity demand and contributed 0.3% of all carbon dioxide emissions in 2020. A typical data center uses 3 million to 5 million gallons of water per day, equivalent to the water consumption of a city of 30,000 to 50,000 people.

Microsoft has previously said it plans to run all of its data centers on 100% renewable energy by 2025. Meta claims to have achieved this feat in 2020.

The above is the detailed content of Global IT giants are using AI to help data centers save energy and reduce emissions.. For more information, please follow other related articles on the PHP Chinese website!