① Refined energy consumption management
Currently, server chips can reduce energy consumption by shutting down some functions when idle, and adjustable server fan speeds are widely used. These technologies facilitate the linkage and unified energy reduction between L1 layer power distribution equipment and L2 layer IT equipment. With the widespread application of sensors, local environmental and power consumption data within data centers can be easily obtained. A comprehensive energy management system that integrates the internal physical ecosystem (power distribution system, temperature control system, IT equipment, etc.) and the external natural ecosystem (power grid information, meteorological information, user usage information, etc.) of the data center is beginning to take shape. Based on this, some internet giants will use machine learning technology to ultimately achieve self-optimization and watt-level energy management of the entire data center.
② Information processing and value mining
As a network element, UPS will work in conjunction with distribution cabinets and air conditioning. As the medium between the power grid and the equipment, and as one of the managed network elements, the power supply and distribution system, including UPS, will undertake more information collection and processing functions. Power grid data and load data will be valuable information in the future. Through data analysis, it will create a comprehensive power consumption map for the data center, allowing customers to quickly access relevant information such as the most power-consuming servers and the cyclical patterns of power consumption. Furthermore, the application of internet big data technology will further unlock the value of this information. Some internet companies have already begun researching this, and it is believed that in the near future, power supply and distribution systems, primarily UPS systems, may transform from purely consumable equipment into profitable production equipment.
③ Application of Internet of Things (IoT) technology
With the further popularization of mobile communication technology, the era of the Internet of Things has arrived. Network management apps, mobile operations and maintenance, and automated maintenance will gradually replace traditional operation and maintenance methods. Imagine a data center located near a solar power station in Inner Mongolia, with no one in the server room. Thousands of miles away in Beijing, an IT maintenance worker from a company checks their mobile phone and finds an alarm from the network management system: “UPS #12 load has exceeded expectations.” After automatic tracking, the system finds that the power consumption of several servers has increased significantly, due to a surge in the number of users of a game running on the servers. The system then offers two options: “1. Activate the backup power supply module (increases system electricity costs by 65 yuan/day)” and “2. Migrate some applications to idle server #453 (increases system electricity costs by 32 yuan/day).” The network management system then automatically activates the power supply module or server according to the selection. In addition to the daily management of data centers, the application of IoT technology will also bring about a revolution in equipment maintenance methods, such as equipment life prediction, fault prevention, and fault handling. Users can purchase the corresponding maintenance products according to the prompts of the network management system, just like shopping on Taobao, and remotely authorize the manufacturer’s maintenance personnel to go to the fault location for maintenance. The single-board-level spatial positioning system can guide the maintenance personnel to quickly replace the faulty components.