Skip to content
All tool news

Tool desk · Updated daily

Tool news·Vercel·

Vercel Groups CDN Metadata Lookups to Cut P99 Latency 91%.

Vercel now groups CDN metadata lookups, cutting P99 latency by 91% and reducing cache misses across large deployments.

CW

Create With tool desk

3 sources checked · 2 min read · 3 sections

ShareLinkedIn
Vercel Groups CDN Metadata Lookups to Cut P99 Latency 91%

Vercel’s grouped CDN metadata lookups have cut P99 lookup latency by 91%. The change replaces one-path-at-a-time fetching with grouped requests, reducing cache misses for large deployments and improving deployment speed.

According to the engineering blog post, every request to Vercel passes through its CDN. That network executes more than 80 million routing instructions per second on average. Part of this work involves finding metadata that tells the CDN which paths exist and how each one should be served.

P99 metadata lookup latency
91% lower
after production experiments
Routing instructions
Over 80 million per second
CDN average
Deployment size
Hundreds of thousands of paths
possible on large sites

When the required metadata is not cached, the CDN must fetch it before serving the response. Vercel previously fetched and cached metadata for one path at a time. Each lookup retrieved only the information it needed.

Grouped CDN Metadata Lookups Reduce Cache Misses

Fetching one path at a time created a large number of separate cache entries. A deployment containing hundreds of thousands of paths could have an entry for every path. Each new deployment then introduced fresh metadata, making cache misses a recurring cost for large sites that deployed frequently.

The updated system groups paths together. Fetching metadata in groups means one cache operation can cover more than one path, rather than repeating the process separately for each path. Vercel ran production experiments to find a balance that reduced cache misses without replacing the original problem with excessive fetching.

CDN metadata processPrevious approachUpdated approach
FetchingOne path at a timePaths grouped together
Cache entriesOne for each pathShared across grouped paths
P99 lookup latencyBaselineReduced by 91%

The resulting configuration cut P99 metadata lookup latency by 91%. It also improved deployment speed, addressing the repeated metadata cost created when large sites published new deployments.

What Builders Need to Change

This is an infrastructure optimisation rather than a new dashboard control or deployment setting. Existing requests still pass through Vercel’s CDN, while the metadata lookup process behind those requests now uses grouped paths.

The work follows another infrastructure performance update that made Vercel Sandbox routing 18x faster worldwide.

Frequently asked questions

What are Vercel grouped CDN metadata lookups?

Grouped CDN metadata lookups fetch information for multiple paths together. Vercel previously fetched and cached this metadata one path at a time.

How do grouped lookups compare with the previous approach?

The previous system could create a separate cache entry for every path in a deployment. Grouping paths reduced recurring cache misses and cut P99 metadata lookup latency by 91% in production experiments.

When are grouped CDN metadata lookups available?

Vercel reported results from production experiments in a blog post dated 10 September 2026. The post does not give a separate rollout date or plan availability.

Sources

3 checked

How we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.

Worth passing on?

ShareLinkedIn

Go deeper on Vercel

Related reading, watching and going.

Everything on Vercel →

The briefing

19 May 2026

How to Migrate Your App from Lovable to Claude Code in 2025

Step-by-step guide to migrating your Lovable app to Claude Code. Learn how to move from Lovable's managed platform to a stack you own and control.

Read →

The briefing

10 Feb 2026

OpenClaw Review: 2 Weeks With This AI Agent

Honest 2-week review of OpenClaw — an autonomous AI agent powered by Claude. We test setup, safety, autonomy, and real value to see if it's worth using.

Read →
34 min

Podcast

My Daughter Built a Website Business, Claude Design First Look & Going All-In on Cloudflare

James and Kieran discuss practical AI-powered development through real-world examples, including a seven-year-old building a website business with Claude Code voice mode, Anthropic's new Claude Design tool, API security challenges in the AI agent era, and Cloudflare's expanding AI infrastructure offerings. The episode covers hands-on experiences with vibe coding, emerging design tools, and the evolving landscape of AI-assisted development.

Watch now ↗

Podcast

53. Why Your Vibe Code Needs Tests, Ditching SaaS & AI Agents That Work While You Sleep

James and Kieran discuss the practical realities of building AI-powered software, emphasizing the importance of testing in vibe-coded applications, exploring alternatives to traditional SaaS tools, and implementing autonomous AI agents for business operations. The episode covers technical topics like Convex databases, app store bottlenecks from AI-generated submissions, and using OpenClaw as a marketing automation agent.

Watch now ↗
38 min

Podcast

52. Your AI Co-Founder: OpenClaw, Manus & the Zero-Person Business

In this Create With podcast episode, James and Kieran are joined by Matt Roberts from Happy Operators for an in-depth discussion on using autonomous AI agents in production. The conversation covers practical experiences with OpenClaw, security considerations, comparison with emerging tools like Manus, and the broader shift toward AI-assisted development. They also discuss the VibeCoding Olympics results, Bubble's new AI capabilities, and research on Claude.MD file effectiveness.

Watch now ↗

Latest tool news

What else changed this week.

All tool news
VercelDigest

What Vercel Shipped in Late October

Vercel shipped quicker static deployments, four AI Gateway models, more free domain choices and new CLI controls in late October. The changes cover developers testing sites…

Vercel

Vercel Adds Submit_feedback to Its MCP

submitfeedback is a new Vercel MCP tool that lets agents and harnesses send bug reports, friction reports and feature requests directly to the Vercel team. Feedback can include…

The Create With Briefing

Don't watch forty changelogs. Read one email.

Every Tuesday: the tool changes worth knowing, real business use cases, and what's on near you. Free, unsubscribe any time.