整理 | 郑丽媛 转自 | CSDN（ID：CSDNnews） 投稿或寻求报道 | zhanghy@csdn.net 你是不是也有几部旧手机尘封在抽屉深处？ 可能 屏幕 有些 破碎 、电池 不太行 ， 抑或 系统 卡顿 、性能落后 ， 总之 早已 退役 —— 而 现在，它们 或许 能"重返一线" 了 。 在 这个 "AI Everywh ere" 的 时代，算力就是生产力 ， 但 并非 所有场景都适合部署昂贵、能耗高 的大型数据中心。 于是 ， 近期 一项来自 塔尔图大学 （University of Tartu） 的 研究 指 出： 你 手里闲置 的 那 几部 旧 手机 ， 其实 就 是一座座 等 待激活的"微型数据中心" 。 实验显示， 哪怕 是 几部 10 年 前 的 老 手机， 都能 高效完成图像识别、传感器数据处理等任务 。 更重要的是 ， 整体 改造成本 只需 要 8 欧元 （ 约合 人民币 6 6 元 ） ！ 为什么选择旧手机 ， 它们真的"能打"吗？ 你可能会觉得，用智能手机搭建数据中心听起来有点不靠谱。 毕竟 ， 这些旧手机电池早就不耐用 了，操作系统也面临大量安全漏洞。 因此 许多人 都认 ...

Core Insights - The research from the University of Tartu suggests that old smartphones can be repurposed as micro data centers, providing a cost-effective solution for distributed computing needs in the era of AI [1][4][15] Group 1: Research Findings - Even smartphones from a decade ago can efficiently perform tasks such as image recognition and sensor data processing, with a total modification cost of only 8 euros (approximately 66 RMB) [3][10] - The study highlights that smartphones are designed for high-energy scenarios, making them suitable for complex computational tasks despite their outdated hardware and software [3][4] - The research team successfully transformed four old Google Nexus phones into a micro data center, demonstrating the feasibility of this approach in real-world applications [5][12] Group 2: Methodology - The transformation process involved removing the original battery and using an external power supply to ensure reliable operation [7] - The original Android operating system was replaced with PostmarketOS or similar lightweight Linux-based systems to enhance control and flexibility [8] - A cluster of four smartphones was created, with one acting as the master node to manage tasks and the others as worker nodes for parallel processing [9] Group 3: Applications and Implications - The micro data center was tested in an underwater experiment, successfully identifying and counting marine life over an 8-hour period without failure [10][12] - Potential applications include urban sensor networks for crowd counting and providing real-time image analysis for drones [12] - The research indicates that this approach could benefit startups that require data analysis and web hosting services without the high costs associated with cloud computing [12][15] Group 4: Environmental Impact - The study addresses the growing issue of electronic waste, with approximately 62 million tons generated globally in 2022, of which only 22% was properly recycled [14] - Repurposing old smartphones as reusable computing resources presents a viable solution to reduce electronic waste and its environmental impact [14][15] - The research encourages manufacturers to consider maintainability and reusability in product design to mitigate electronic waste generation [15]