展示HN：GizmoEdge – 分布式物联网SQL引擎
你好，
我是菲利普·摩尔（Philip Moore），GizmoData的创始人，以及GizmoEdge的创造者——一个由物联网（IoT）设备驱动的分布式SQL引擎。
GizmoEdge是一个原型应用程序，允许你运行SQL查询，这些查询会分发到多个设备上，包括：Linux、macOS、iOS、iPadOS、Kubernetes Pods、树莓派等。
我构建了一个前端应用程序，你可以通过它发出分布式SQL查询，网址是： [https://gizmoedge.gizmodata.com](https://gizmoedge.gizmodata.com)
如果你有苹果设备，可以安装GizmoEdge Worker应用程序，让它“加入集体”，以下是App Store链接：[https://apps.apple.com/us/app/gizmoedge/id6738658135](https://apps.apple.com/us/app/gizmoedge/id6738658135)
安装应用后，你只需将其连接到正在运行的GizmoEdge服务器，它将下载一个1GB大小的TPC-H数据分片（实际上更小，因为它是压缩的parquet格式，存储在ZStandard压缩的tar包中）。然后，该应用会在你的设备上从parquet数据集创建一个DuckDB数据库。
当你在GizmoEdge SQL应用（[https://gizmoedge.gizmodata.com](https://gizmoedge.gizmodata.com)）中发出符合分发条件的SQL查询时，它应该会在你的设备（以及其他设备）上进行分发和执行。
如果大家能尝试这个引擎，我将感到非常荣幸。请记住，这只是一个原型，目前还未准备好投入生产使用。它目前也是只读的，但我正在努力寻找一种方法，以便在不久的将来让大家更容易地将数据导入以进行分布式SQL执行。
GizmoEdge在工作节点上由DuckDB提供支持，并使用WebSockets实现服务器与工作节点之间的低延迟连接。它使用TLS加密通信，并拥有一个强大的安全模型，其中服务器和工作节点之间存在“信任但需验证”的关系。
开始使用：
GizmoEdge SQL导航应用（在此运行交互式SQL查询）：[https://gizmoedge.gizmodata.com](https://gizmoedge.gizmodata.com)
GizmoEdge Worker在App Store上的链接：[https://apps.apple.com/us/app/gizmoedge/id6738658135](https://apps.apple.com/us/app/gizmoedge/id6738658135)
GizmoEdge主页：[https://gizmodata.com/gizmoedge](https://gizmodata.com/gizmoedge)
感谢你的时间和反馈！
查看原文
Hi,<p>I'm Philip Moore - the founder of GizmoData, and creator of GizmoEdge - a Distributed SQL Engine powered by Internet-of-Things (IoT) devices.<p>GizmoEdge is a prototype application - that allows you to run SQL queries that get distributed to many devices - including: Linux, macOS, iOS, iPadOS, Kubernetes Pods, Raspberry Pis, and more.<p>I've built a front-end application from which you can issue distributed SQL queries - here: <a href="https://gizmoedge.gizmodata.com" rel="nofollow">https://gizmoedge.gizmodata.com</a><p>If you have an Apple device - you can install the GizmoEdge Worker app to have it "join the collective" - here is the App Store link: <a href="https://apps.apple.com/us/app/gizmoedge/id6738658135">https://apps.apple.com/us/app/gizmoedge/id6738658135</a><p>Once you install the app, you just connect it up to the running GizmoEdge server - and it will download a shard of TPC-H data at the 1GB foot-print (it is actually smaller, b/c it is compressed parquet - in a ZStandard-compressed tarball). The app will then create a DuckDB database from the parquet datasets that is on your device.<p>When you issue a SQL query that is eligible for distribution from the GizmoEdge SQL app (at <a href="https://gizmoedge.gizmodata.com" rel="nofollow">https://gizmoedge.gizmodata.com</a>) - it should distribute and execute on your device (among others) in the collective.<p>I would be honored if folks tried the engine. Please bear in mind - this is a prototype - it isn't production ready, yet. It is currently read-only as well - but I'm working on a way to make it easy for folks to ingest data for distributed SQL execution in the near-future.<p>GizmoEdge is powered by DuckDB on the workers, and uses WebSockets for low-latency connections between the server and workers. It uses TLS for encryption of communication - and has a robust security model - in which the server and workers have a "trust-but-verify" relationship.<p>Getting started:
GizmoEdge SQL Navigator app (run interactive SQL queries here): <a href="https://gizmoedge.gizmodata.com" rel="nofollow">https://gizmoedge.gizmodata.com</a>
GizmoEdge Worker on the App Store: <a href="https://apps.apple.com/us/app/gizmoedge/id6738658135">https://apps.apple.com/us/app/gizmoedge/id6738658135</a>
GizmoEdge homepage: <a href="https://gizmodata.com/gizmoedge" rel="nofollow">https://gizmodata.com/gizmoedge</a><p>Thank you for your time and feedback - in advance!