TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Ask HN: Does a dynamically scaling cloud workstation exist somewhere?

7 点作者 ekns超过 3 年前
I frequently work with some &#x27;data science&#x27; projects that go from some MB to hundreds of GB.<p>Ideally I&#x27;d have a cloud terminal I&#x27;d connect to which could scale its RAM to fit my process RAM usage (and possibly scale up CPUs transparently too).<p>I know that you can scale up various cloud instances, but managing the runtime state is a problem. I&#x27;d like to avoid ever having to kill whatever processes I have running.<p>Something like Google&#x27;s Live Migration would also be a good match here, if it enabled migrating to a bigger machine type without rebooting, or without otherwise losing process state.<p>Ideally I&#x27;m looking for something that I could transparently scale up and down, and which I could always SSH into without having to manually start&#x2F;shutdown the instances.<p>Bonus points if GPUs could be added&#x2F;removed in the same manner.

1 comment

tgdn超过 3 年前
Have you looked into Spark? There are managed Spark options on AWS&#x2F;GCP (for example Databricks). Spark lets you do exactly what you are saying.<p>Define minimum&#x2F;maximum number of nodes, the machine capacity (RAM&#x2F;CPU) and let Spark handle the scaling for you.<p>It gives you a Jupyter-like runtime to work on possibly massive datasets. Spark is perhaps too much for what you&#x27;re looking for. Kubernetes could possibly be used with Airflow&#x2F;DBT possibly, for example for ETL&#x2F;ELT pipelines.
评论 #28855011 未加载