展示HN：将指标、日志、追踪和事件集中于一处。这就是Arc。
嗨，HN，
我之前在这里提到过Arc，现在想分享一个快速更新。过去几周我们一直在全力以赴，进展迅速，性能提升，功能逐渐落地，社区反馈非常积极。
Arc使用DuckDB作为SQL引擎，底层支持MessagePack、Arrow和Parquet。目前，我们的指标大约达到245万RPS，日志、事件和追踪大约为100万RPS。现在它支持保留策略、删除操作以及随着数据到来进行聚合的连续查询。
我们还开发了一个VS Code扩展，用于探索和查询Arc，增加了Apache Superset集成，并且正在着手开发Telegraf和Grafana数据源。
欢迎任何反馈，无论是好是坏，甚至是严厉的批评，都会对我们有所帮助。
祝您周四愉快！
查看原文
Hey HN,<p>I’ve posted here before about Arc, and wanted to share a quick update. We’ve been heads down the last few weeks, and things are moving fast, performance is up, features are landing, and the community response has been awesome.<p>Arc uses DuckDB as the SQL engine, with MessagePack, Arrow, and Parquet under the hood. Right now we’re seeing around 2.45M RPS for metrics, and about 1M RPS for logs, events, and traces. It now supports retention policies, deletes, and continuous queries that aggregate data as it comes in.<p>We also built a VS Code extension to explore and query Arc, added Apache Superset integration, and are working on Telegraf and a Grafana datasource next.<p>Feedback is always welcome. good, bad, or brutal. It all helps.<p>Happy Thursday!