潜伏

数据可观测性初创企业营收为何迅猛增长_我的网站

一 |     

    This photo taken on July 10, 2026 shows a data-driven environmental governance system, known as the innovative regulation-monitoring-inspection linkage model, showcased at the Beijing Municipal Ecological and Environmental Monitoring Center in Beijing, capital of China. (Xinhua/Sun Suying)Beijing's blue skies have become a visible sign of the city's progress in tackling air pollution, and behind that progress, digital technologies are playing an increasingly important role.
In 2025, Beijing recorded 311 days with good air quality and only one heavily polluted day, compared with 58 heavily polluted days in 2013, when the city officially began releasing PM2.5 monitoring data. The annual average PM2.5 concentration, a key indicator of air quality, fell from 89.5 micrograms per cubic meter in 2013 to 27 micrograms per cubic meter in 2025, a decline of nearly 70 percent.
The progress has drawn international attention. The United Nations Environment Programme has highlighted Beijing's experience in improving air quality as a model for other cities.
According to experts at the Beijing Municipal Ecological and Environmental Monitoring Center, the cleaner air is the result of a comprehensive system integrating environmental monitoring, scientific analysis, policymaking, regional cooperation and law enforcement -- with digital technologies now woven into every link of that chain.
MORE TARGETED APPROACH
Effective governance begins with accurate monitoring. Beijing has established a multi-level air-quality monitoring network. Figures show 35 standard monitoring stations across the city continuously measure six major pollutants, while more than 1,000 compact sensors deployed across communities and townships provide more detailed information on local air-quality variations.
The monitoring system was gradually built by the Beijing Municipal Ecological and Environmental Monitoring Center, one of China's earliest ecological and environmental monitoring institutions, responsible for monitoring Beijing's air, water, noise, soil environments and ecosystems.
The center has developed an analytical system that can identify around 125 components of PM2.5, covering most of its composition and helping researchers trace pollution sources.
"Source analysis helps us identify the major contributors to pollution at different stages, providing a scientific basis for policymakers to target key emission sources," said Li Yunting, technical director at the center.
Coordinated by the Beijing Municipal Ecology and Environment Bureau, the center has conducted four rounds of PM2.5 source analysis studies with other institutions, with results released in 2014, 2018, 2021 and 2025.
The latest analysis showed that regional transport contributed 57 percent of Beijing's PM2.5 pollution. Among local sources that accounted for the remaining 43 percent, vehicle emissions and residential sources represented the largest share of pollution, a distinctive characteristic of a major international metropolis.
The monitoring results have provided scientific support for a series of pollution-control measures. For example, Beijing has promoted new-energy vehicles to reduce transport emissions and adopted advanced techniques, including air-supported membrane structures, to control dust emissions. Meanwhile, coal combustion pollution, once a major contributor, has almost been eliminated and is no longer listed among the main sources, reflecting broader national efforts to reduce coal consumption, promote clean energy and upgrade heavily polluting industries.
With PM2.5 concentrations now at relatively low levels, further gains in air quality will depend on more targeted, science-driven approaches.
    This photo taken on July 10, 2026 at the Beijing Municipal Ecological and Environmental Monitoring Center in Beijing, capital of China, shows the changes in the annual average concentration and spatial distribution of PM2.5 in Beijing from 2013 to 2025. (Xinhua/Sun Suying)
SMARTER ENVIRONMENTAL GOVERNANCE
Beyond monitoring pollution, Beijing has built a data-driven environmental governance system that links monitoring, regulation and enforcement in a closed loop from detection to action and follow-up evaluation.
Developed by the center and known as the innovative regulation-monitoring-inspection linkage model, the system integrates artificial intelligence, the Internet of Things, satellite remote sensing and vehicle-mounted monitoring technologies to pull together environmental data from multiple sources onto a unified platform.
"The system brings information that was once scattered across different departments onto a single platform," said Zhang Likun, director of the smart system division at the center, adding that this enables off-site enforcement and joint cross-sector actions involving housing, water resources and transportation, significantly improving the efficiency and precision of environmental management.
Using more than 30 intelligent algorithms, the platform automatically analyzes data and identifies potential pollution risks, based on which monitoring staff identify the most pressing problems and forward them to enforcement departments.
The approach has already drawn international attention. Clean Air Asia, an international nongovernmental organization focused on improving air quality across Asia, featured the system as a representative case study in its "China Air 2025" report. The model was also presented at the 2026 Better Air Quality Conference and recognized with the Certificate for Clean Air Best Practice Sharing.
China has integrated ecological conservation into its modernization drive. As the country speeds up efforts under the Beautiful China Initiative, Beijing's experience demonstrates how technology can support more precise, coordinated and sustainable environmental governance.
。    　　作者：阿历克斯・库图尔　　一批向 Snowflake、Databricks、Datadog 等数据库巨头发起挑战的初创企业，正借着 AI 智能体普及的东风迎来营收暴涨。这或将催生一波初创企业交易浪潮，涵盖新一轮融资以及兼并收购。　　本周二早些时候，我报道过 ClickHouse 的年度经常性收入再创新高；这家企业提供免费开源软件与付费云服务，帮助企业分析海量数据，OpenAI 今年对其产品的使用量大幅攀升。　　OpenAI 的动向颇具代表性：这家 ChatGPT 开发企业同时也是上市公司 Datadog 的客户。Datadog 是可观测性领域的行业巨头，可监控人与 AI 智能体在计算机上的各类行为。据知情人士透露，ClickHouse 上一轮融资发生在今年 1 月，估值达 150 亿美元，今年已经收到多家战略收购方的接洽邀约。　　首席执行官亚伦・卡茨向我表示：“我们不会无视外界的收购意向，毕竟要对股东负责，但我们不会出售公司，恰恰相反，我们还要继续独立发展。”　　受益于 AI 智能体带来的海量数据需求的并非只有 ClickHouse。Cribl 主营遥测技术，帮助企业迁移、存储用于监控的各类数据，该公司称其年度经常性收入接近 4 亿美元，较 2 月份上涨 33%。　　Cribl 联合创始人兼首席执行官克林特・夏普表示：“我们的企业客户正面临数据规模爆发式增长，部分动因来自人工智能以及智能体的兴起，而我们正在帮客户应对这一局面。”　　这家成立已有八年的企业服务 Zoom、ServiceNow、希尔顿等客户，帮助客户把数据分流到成本更低的存储介质，同时匹配最高效的监控工具。　　夏普称：“我们解决的现实痛点是，企业数据以30% 的复合增速持续膨胀，但 IT 预算却没有同步增长。”　　Cribl 上一轮风投融资是 2024 年 8 月，估值 35 亿美元。夏普透露，公司计划未来两三年内启动 IPO 上市。　　拥有 12 年历史的 Grafana，主打工具让开发者实时可视化查看应用、云服务器与 AI 智能体的运行状态，近期营收同样迎来提速。

二 | Grafana 首席营销官斯科特・芬格胡特表示，部分增长来源于自家 AI 助手，该助手可以帮助企业监控自有应用的运行活动。　　芬格胡特介绍，今年有半数 Grafana 客户使用这款 AI 助手，完成系统监控配置与工具调试。　　该公司去年 9 月年度经常性收入突破 4 亿美元。投行软件行业人士表示，Grafana 经常被视作大型企业潜在的收购标的，那些觊觎数据监控赛道的巨头都有可能出手。　　Elastic、Snowflake 接连出手收购　　大型企业已经开始收购专注数据监控赛道的初创公司。　　提供开源大数据搜索引擎的上市公司 Elastic，于周一斥资 8500 万美元收购 Deductive AI，该产品可以协助工程师监控数据、排查故障。两周前，Dynatrace 花费 9.15 亿美元收购 Arize AI，这家公司专门追踪 AI 模型与智能体的运行表现。　　Snowflake 在 6 月完成 10 亿美元收购 Observe 的交易，押注已经把数据存放在自家平台上的客户，会需要配套工具查看应用运行性能。在此笔收购之前，网络安全企业帕洛阿尔托网络以 33.5 亿美元收购可观测性创业公司 Chronosphere。　　投资人与投行人士判断，后续还会出现更多收购。网络安全厂商、谷歌、亚马逊、微软等科技巨头，还有思科这类网络设备厂商，都有可能入局收购数据监控类初创企业。　　Founders Circle Capital 合伙人迈克・荣格表示，该机构投资过 Databricks、Vercel、ClickHouse、Cribl。他称：“随着 AI 智能体承担越来越多自主工作，企业需要看清智能体到底在做什么，并且建立防护机制，在故障、安全风险造成实际损害之前及时拦截。”责任编辑：郭明煜。

Current article:http://ycwxn.feishuozhuangkeqi.cyou/list_smwg/v8q34u3.html

Published on:10:33:07