Uptime is a measure of a service's or system's availability, both in the short and long term. Organizations prioritize high uptime for their services — and especially so in sectors such as banking, healthcare, and government — since it supports good user experiences and access to critical personal identifiable information (PII). This is typically measured with a simple percentage, using the formula (total time - downtime) / total time x 100.

How does uptime work?

Uptime is one of many factors used to evaluate quality of service (QoS) of an application, API, or LLM. Uptime is also inversely related to downtime — or the measure of how often a service is unavailable. More uptime means less downtime, and vice versa. Organizations commit many resources (both technical and human) to achieving 99.999% availability for their services, which many view as the realistic pinnacle of uptime. 

That said, uptime carries diminishing returns past a certain point. What this threshold is depends on an organization's unique infrastructure, resources, and overall goals. It's relatively easy to take a vulnerable, crash prone service and make it somewhat dependable. However, making the jump from 95% uptime to 99%+ uptime (the "last mile" of conventional high availability) typically requires immense planning, experimentation, and investment. 

Companies should understand what degree of uptime they and their users require versus what they should aspire to internally. For example, a company might declare, "We only want to see a few hours of monthly downtime" whereas others may only accept minutes of monthly downtime. The criteria determining whether a piece of infrastructure is working as designed is subjective — while often tied to user expectations and profitability, as we've seen in eCommerce. Low priority services are generally given more leeway as any impacts won't be felt as quickly or immensely. 

That said, maintaining uptime requires the following: 

  • Continuous monitoring, end-to-end observability, and visibility into infrastructure — enabling teams to see what's working and what isn't while keeping tabs on traffic patterns

  • An automated alerting system — letting key DevSecOps team members know when outages or unforeseen issues strike

  • Detailed logging mechanisms that can capture runtime data, user activity, and reported errors over critical time periods

  • Multi-layered security systems that can identify and block bad traffic and stop cyber attacks — with few false positives and false negatives

  • Efficient, well-integrated systems that prioritize flexibility over lock-in

  • High availability features such as GSLB, failover, VRRP, disaster recovery, health checking, and graceful routing changes (blue/green deployments, canary deployments, etc.)

You’ve mastered one topic, but why stop there?

Our blog delivers the expert insights, industry analysis, and helpful tips you need to build resilient, high-performance services.

By clicking "Get new posts first" above, you confirm your agreement for HAProxy to store and processes your personal data in accordance with its updated Privacy Policy, which we encourage you to review.

Thank you! Your submission was successful.

Can HAProxy help boost uptime?

Yes! High availability (HA) is in our DNA and is a core goal across our products. As the world's fastest application delivery and security platform, HAProxy One builds on HAProxy's performance with HAProxy Enterprise WAF, HAProxy Enterprise Bot Management Module, Global Rate Limiting, and other advanced, multi-layered security features to help you get closer to 99.999% uptime. 

Plus, functions like failover protection, retries, health checking, and traffic overload protection (request queueing) help you prevent disasters while boosting uptime. To learn more, check out our high availability page or our HAProxy Enterprise datasheet.