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。

Why Vercel overhauled its serverless infrastructure for the AI era

2 点作者 sylvainkalache3 个月前

1 comment

nadis3 个月前
Interesting shift from speed to cost-savings and from web app development to AI-native apps.<p>&quot;But as Vercel&#x27;s customers started using the serverless platform to build AI apps, they realized they were wasting computing resources while awaiting a response from the model. Traditional servers understand how to manage idle resources, but in serverless platforms like Vercel&#x27;s &quot;the problem is that you have that computer just waiting for a very long time and while you&#x27;re claiming that space of memory, the customer is indeed paying,&quot; Rauch said.<p>Fluid Compute gets around this problem by introducing what the company is calling &quot;in-function concurrency,&quot; which &quot;allows a single instance to handle multiple invocations by utilizing idle time spent waiting for backend responses,&quot; Vercel said in a blog post last October announcing a beta version of the technology. &quot;Basically, you&#x27;re treating it more like a server when you need it,&quot; Rauch said.<p>Suno was one of Fluid Compute&#x27;s beta testers, and saw &quot;upwards of 40% cost savings on function workloads,&quot; Rauch said. Depending on the app, other customers could see even greater savings without having to change their app&#x27;s configuration, he said.&quot;