Edge Computing and the Need for Low Latency

边缘计算与低延迟场景

课文音频

可先听下面的整段朗读；需要逐句点读时，请打开阅读器，体验会更顺手。

课文

When self-driving cars detect a pedestrian, they must react in under 100 milliseconds — too slow if data travels to a distant cloud server.

自动驾驶汽车检测到行人时，必须在100毫秒内做出反应——若数据需传至遥远的云服务器，则响应过慢。
Edge computing places processing power close to where data is generated, such as inside traffic cameras or factory machines.

边缘计算将算力部署在数据生成地附近，例如交通摄像头或工厂设备内部。
This reduces delay, or latency, by cutting transmission distance and avoiding network congestion during peak hours.

此举通过缩短传输距离、避开高峰期网络拥堵，显著降低延迟。
In smart factories, edge devices instantly adjust robotic arms when sensors spot misaligned parts on a conveyor belt.

在智能工厂中，边缘设备一旦通过传感器发现传送带上零件错位，便立即调整机械臂。
Video analytics for crowd monitoring also relies on edge AI to flag anomalies without sending raw footage online.

人群监控的视频分析同样依赖边缘AI，可直接识别异常，无需上传原始视频。
Unlike traditional cloud models, edge systems prioritize speed and local decision-making over centralized storage.

与传统云模型不同，边缘系统更注重速度和本地决策，而非集中式存储。
They often work alongside the cloud: edges handle urgent actions, while the cloud analyzes long-term trends.

边缘系统常与云协同工作：边缘处理紧急任务，云端分析长期趋势。
5G networks further boost edge computing by providing ultra-fast, reliable connections between devices and local servers.

5G网络凭借超高速、高可靠的设备与本地服务器连接，进一步推动边缘计算发展。