Operations | Monitoring | ITSM | DevOps | Cloud

In today’s world of globally distributed applications, user experience is everything. Whether your platform runs across multiple cloud providers or uses a Multi CDN with numerous points of presence (PoPs), efficiently routing user traffic can make or break performance. That's where intelligent traffic steering becomes not just a nice-to-have, but a must-have.

We hosted the first IBM and Catchpoint Retail Digital Performance event on Wednesday, March 19, 2025. The sessions offered practical, thought-provoking insights on speed, resilience, and user-centric design—giving attendees fresh strategies to improve digital experiences at scale.

While incidents like last year’s CrowdStrike outage tend to dominate headlines, far more often, the real battle for Internet Resilience isn’t fought on a global stage. It’s waged in the shadows of financial districts, within overloaded cloud data centers, or a rural ISP’s overtaxed peering points. Traditional monitoring tools, designed for broad strokes, miss these hyper-specific failures.

On March 20, 2025, starting at 15:43 AM UTC, Zendesk users globally encountered 503 “Service Unavailable” errors and 5xx server-side issues, disrupting access to critical support tools and communication channels. While immediate mitigations stabilized core services, intermittent issues continued for over 24 hours, underscoring the complexity of multi-pod infrastructure failures.

Slow is officially the new down. That’s a major finding of The SRE Report 2025, with 53% of study respondents agreeing with this expression, and 44% stating “performance should be tracked against a service-level objective.”

When X (formerly Twitter) suffered a global outage on March 10-11, 2025, millions of users and businesses were left in the dark. Apart from a solitary post from CEO Elon Musk claiming a cyber-attack, X has remained silent. Yet Catchpoint’s Internet Sonar detected the crisis in real time—highlighting the critical role independent, proactive monitoring plays when vendor communication fails.

In the early days of DoubleClick, back when SaaS was still known as Application Service Provider (ASP), I was tasked with setting up the QoS (Quality of Service) Team. Our primary mission was to establish a monitoring system, but we quickly found ourselves managing Service Level Agreements (SLAs)—a task that became critical after we paid out over $1 million in penalties for SLA violations to a single customer. The reason? Someone had signed a contract promising 100% uptime, an impossible commitment.

AI platforms have experienced several service interruptions over the past few months. We’ve all seen the memes fly when ChatGPT, Gemini or Perplexity go down. They’re funny at first, but then reality hits: if you rely on AI tools for work or business, these outages can grind your day to a halt.