Zendesk

99.999% Uptime and 50% Latency Reduction in 5 Weeks

How Zendesk modernized its global network infrastructure and introduced GitOps practices with an embedded EverOps Pod.

99.999%
uptime for remote access globally
50%
reduction in network latency

Zendesk is a cloud-based industry leader in help desk management, providing customizable tools for customer service portals, knowledge bases, and online communities. With a service-first CRM approach, Zendesk serves millions of customers worldwide.

EverOps has partnered with Zendesk on initiatives ranging from office build-outs and data center retirements to Kubernetes adoption—a relationship now entering its second decade.

The Challenge

A Global Network That Couldn't Scale

Zendesk operates an IT network supporting over 5,000 users across five continents. As traffic volume and use case complexity grew, the network's maintenance process became a liability—difficult to scale and slow to recover from outages.

The team faced several compounding issues: fragmented monitoring across multiple platforms made it hard to get a holistic view of network health. Integrating data from various sources required significant manual effort. Different systems had inconsistent alerting thresholds, delaying issue resolution. And a lean IT staff was stretched thin managing tools while handling incidents.

Leadership needed to reduce network latency across all global regions, improve overall availability, and dramatically cut mean time to recovery (MTTR)—without adding headcount.

The Approach

Embedded Engineers with a Phased Roadmap

EverOps embedded directly with Zendesk's leadership and engineering teams over several weeks to assess the current state of network architecture and traffic patterns. Rather than delivering a report and walking away, our engineers worked inside Zendesk's environment to design and execute a multi-phased transformation.

The engagement targeted three functional areas: VPN configuration management, DNS infrastructure, and global traffic optimization. Each phase was scoped for rapid delivery with clear milestones, allowing the team to validate improvements incrementally while maintaining production stability.

Throughout, EverOps operated as an extension of the internal team—collaborating in real-time, executing changes directly, and transferring knowledge to ensure Zendesk could maintain and extend the solution independently.

Technologies & Tools: Palo Alto Panorama · Terraform · Terragrunt · AWS Route 53 · DNS Load Balancing

Technologies & Tools:

No items found.
The Approach

Embedded Expertise with a Clear Roadmap

EverOps embedded a 3-person FinOps Pod directly into Life360's platform team. Within the first week, our engineers had access to their AWS accounts, joined their Slack channels, and began a comprehensive cost analysis using our proprietary Opportunity Finder methodology.

The engagement followed a structured approach: two weeks of discovery and prioritization, four weeks of execution on high-impact optimizations, and two weeks of governance implementation to sustain savings. Throughout, the Pod operated as an extension of the internal team — attending standups, collaborating in real-time, and transferring knowledge continuously.

Unlike consultants who deliver reports, our Pod executed the changes directly — rightsizing instances, eliminating waste, renegotiating reserved capacity, and implementing automated cost controls.

The Solution

Codified Infrastructure with GitOps Best Practices

EverOps delivered a comprehensive network modernization that replaced manual, error-prone processes with automated, repeatable infrastructure-as-code patterns.

Key Initiatives:

  • Panorama Templates for VPN Configuration — Developed variablized VPN config elements across multiple regional firewalls, keeping HA VPN configurations in sync automatically. This eliminated configuration drift and reduced the maintenance burden on the network team.
  • Codified DNS Records in Terraform — Imported 500 existing DNS records through custom scripting and redeployed them via Terraformed modules with Terragrunt, enabling zero-downtime changes and full version control for all DNS configurations.
  • AWS Load Balanced DNS Records — Deployed active-active DNS in geographically disparate regions to optimize global traffic flow and provide seamless, automatic failover between nodes—ensuring users always connect to the closest, healthiest endpoint.
The Results

Enterprise-Grade Reliability in 5 Weeks

Within 5 weeks, EverOps implemented all segments of the solution—including proactive testing and verification—delivering immediate, measurable improvements to Zendesk's global network.

The VPN configuration sync challenges that had plagued the team were resolved through a repeatable, parameterized approach that eliminated configuration errors. GitOps best practices now govern DNS configurations across the entire organization, giving the team full audit trails and rollback capabilities.

Most importantly, Zendesk now has a codified, repeatable network configuration solution that dramatically increases the frequency with which they can release changes—while reducing the effort required. What once took days of manual work and carried significant risk now happens reliably through automated pipelines.

99.999% globally
Remote Access Uptime
50% improvement
Latency Reduction
5 weeks
Time to Implement
500 records
DNS Records Codified
Client Experience

CLIENT EXPERIENCE

—
What's Next?
This engagement represents one chapter in a partnership now entering its second decade. EverOps continues to support Zendesk on infrastructure initiatives—from Kubernetes adoption to ongoing platform modernization.