TechPrufer
Get started
Vercel
Vercel logoVercel · Feature trail

Black Friday Cyber Monday Traffic Operations

10 posts · Sep 2022 – Sep 2026

What this trail covers

Vercel's platform demonstrated its ability to handle massive traffic surges during Black Friday and Cyber Monday 2024, processing over 86.7 billion requests with 99.9992% uptime. Key metrics include 1.9 million peak requests per second, 2.4 million deployments, 3 billion system-triggered Firewall blocks, and 519 million customer-specific WAF blocks. The platform dynamically scaled functions, handled 28.9 billion function invocations, and mitigated 1.75 million DDoS requests per second. It also m

All Vercel trails
  1. Introducing Flat Rate CDNIntroduced Flat Rate CDN for Pro teams, offering predictable monthly billing for CDN usage, including spike protection and tiered capacity for CDN Requests, Fast Data Transfer, Blob Data Transfer, and Observability events. This replaces usage-based billing for these resources, ensuring a fixed monthly cost and preventing performance degradation during traffic spikes. The system monitors usage and rightsizes tiers monthly based on sustained usage.
  2. npx airfry@latest cal.com | VTV CribsThis episode of VTV Cribs visits Cal.com, a scheduling platform, highlighting its office environment and implicitly showcasing the infrastructure that supports its operations. While not detailing specific technical advancements, the visit to a high-growth company like Cal.com serves as a real-world demonstration of Vercel's platform capabilities in handling the demands of such businesses, including scalability and reliability, as evidenced by the company's growth and operational needs.
  3. Billions of requests: Black Friday-Cyber Monday 2025This post details Vercel's performance during Black Friday-Cyber Monday 2025, reporting 115.8 billion total requests (33.6% YoY growth), 518,027 peak requests per second, 6.1 million deployments, 24 million AI Gateway requests, 43.2 billion Fluid compute invocations, 56.9 billion cache hits, 1.8 billion ISR reads and 1.5 billion ISR writes, 7.5 billion firewall actions, and 415 million bots blocked. It highlights the platform's ability to adapt to traffic rhythms across all 20 global regions without manual intervention, emphasizing the seamless integration of AI Gateway, Fluid Compute, ISR, and security systems for handling peak demand.
  4. Update regarding Vercel service disruption on October 20, 2025This post details a service disruption on October 20, 2025, caused by an AWS us-east-1 outage. It describes two independent incidents on the Vercel side: a temporary disruption of production traffic serving due to the AWS data center loss, and a secondary disruption of Vercel's control plane (dashboard, API, builds, log processing) caused by a feature flag provider outage. The post explains the technical flow of a request through Vercel's infrastructure (DNS, CDN, DDoS protection, TLS, Firewall, routing, caching) and identifies the specific points of failure, including a cascading failure in the global caching infrastructure and resource exhaustion in the control plane's Kubernetes cluster. Mitigation steps and restoration timelines are provided for both incidents. The post highlights Vercel's commitment to ownership and accountability for vendor-related issues.
  5. Just another (Black) FridayThis post details Vercel's operational readiness for peak traffic events like Black Friday, highlighting platform capabilities in handling massive request volumes (86.7 billion requests, 1.9 million RPS), security defenses (3 billion platform firewall blocks, 519 million WAF blocks), and intelligent serving strategies (background revalidation, request collapsing). It emphasizes built-in observability features for real-time monitoring of cache hit/miss rates, function invocations, and real-user metrics, along with custom query and anomaly alert capabilities. The post also mentions the ability to stream logs, traces, and analytics to external tools via Drains for unified signal streams. It references a customer success story (Helly Hansen) that saw an 80% YoY revenue increase and zero downtime after migrating to Vercel's CDN and performance optimizations.
  7. Building the Black Friday-Cyber Monday live dashboardThis post details the technical implementation of a real-time Black Friday-Cyber Monday (BFCM) dashboard. It describes the dashboard's modified polling architecture, where the client polls the server every 10 seconds. Backend optimizations include using a rolling window for expensive queries (querying a 5-minute period instead of the entire dataset) and integrating with Upstash KV to store cumulative counts for total metric calculation. Incremental Static Regeneration (ISR) with a 5-second revalidation period was used to cache server responses and reduce database load. Frontend optimizations focused on creating dynamic visuals by calculating the rate of change of metrics and using `requestAnimationFrame` to animate counter increments, providing a smoother user experience than simple polling updates.
  8. Vercel's Black Friday-Cyber Monday 2024 recapThis post details Vercel's performance and operational metrics during Black Friday-Cyber Monday 2024, highlighting the handling of 86.7 billion requests and 1.9 million peak requests per second. It emphasizes the platform's ability to scale functions (28.9 billion invocations), mitigate DDoS attacks (1.75 million requests/sec), and maintain 99.9992% uptime. Specific features like function failover regions, auto-scaling concurrency, routing middleware, and ISR are cited as key enablers for handling traffic spikes, personalization, and ensuring speed. The post also quantifies security measures, including 6.3 billion+ Firewall actions and 519 million custom WAF blocks. It notes an 80% growth in peak requests per second compared to the previous year.
  9. Maximize Uptime with Vercel's Frontend Cloud: How Vercel ensures highly available, globally performant dynamic applications. Even during massive traffic spikes.This post details Vercel's multi-layered approach to frontend resiliency, focusing on how the platform ensures highly available, globally performant dynamic applications, even during massive traffic spikes. It elaborates on optimized global routing at the edge, highly available compute resources (Vercel Functions) with multi-AZ and multi-region redundancy (including Function Failover), and simplified backend failover protection using Edge Config. The post emphasizes that this resiliency comes without additional configuration or maintenance overhead for customers.
  10. Architecting a live look at reliability: Stripe's viral Black Friday site | Customers | VercelThis post details how Stripe built and optimized a real-time microsite to showcase their Black Friday Cyber Monday (BFCM) reliability. Key technical contributions include: architecting a live look at core business metrics, handling 17M+ edge requests at launch with 100% uptime, utilizing Vercel's framework-defined infrastructure to focus on design and reliability rather than cache system architecture, and implementing a strategy combining getStaticProps, Stale-While-Revalidate (SWR) for client-side real-time data and background revalidation, and Incremental Static Regeneration (ISR) with a one-second max-age cache for efficient background data generation and reduced backend load. The post also highlights Stripe's focus on isolation of the microsite's data sources from core infrastructure and a fallback strategy for metric updates.
  11. Managing major traffic spikes during ticket drops with Vercel | Customers | VercelThis post details how Shotgun, an online ticketing platform, leveraged Vercel's serverless autoscaling and Next.js to handle 200x traffic spikes during ticket drops, reducing the need for manual scaling and DevOps overhead. They also utilized Edge Config for rapid waiting room configuration and Vercel Functions for enhanced workflow efficiency. The platform's intuitive UX and UI, along with features like Routing Middleware and Vercel Analytics, support their global expansion efforts.
  12. How the world’s biggest YouTuber served millions of users on VercelThis post details how basement.studio used Vercel to build and scale MrBeast's storefront, handling unprecedented traffic volumes. It highlights the use of Vercel for fast iteration, reliable scaling during peak traffic, and setting a platform record for site requests. The technical implementation involved server-side game logic using Shopify's Admin API for discount codes and a type-safe connection to Shopify's Storefront API via a custom library (react-dropify). Observability was managed with Axiom.