A Brief Overview of Instana Synthetics and GCP uptime Synthetics monitors
GCP Uptime Check — also known as a synthetic monitor — is a feature of Google Cloud Monitoring that tests the availability and performance of applications and resources by simulating user traffic. It periodically sends requests to endpoints (such as HTTP, HTTPS, or TCP services) from multiple global locations and measures the response to ensure the service is up and performing as expected.
Instana Synthetics is a powerful synthetic monitoring feature that proactively tests the availability, performance, and reliability of services by simulating real user interactions. It supports multiple test types, including HTTP/HTTPS uptime checks, API endpoint tests, browser-based transaction flows, and custom scripted scenarios, enabling comprehensive monitoring of application health and user experience.
Instana SRE engineering team successfully executed this initiative - A transformation that not only addressed a critical and time-sensitive business challenge but also utilized and strengthened Instana’s built-in capabilities to deliver scalable, affordable synthetic monitoring without relying on external services.
Context: A Critical Business & Technical Challenge (Strategic Platform Transformation)
Our observability platform depends on continuous uptime monitoring of ~1600 Tenant Units (TUs) to ensure the 99.99% availability that we guarantee to our customers. Historically, this was handled using GCP Synthetics Uptime Check which periodically pings the customers UI endpoint from multiple global locations to measure availability and performance of the TU endpoint.
While effective, this model had fundamental limitations:
- GCP offers only 1 million free checks per month, beyond which every execution is billable.
- Checks cannot be paused — only deleted and defaulted to minimum of 3 locations to run the tests.
- Scaling monitoring across hundreds of TUs resulted in heavy vendor dependency and operational rigidity.
Given our monitoring needs — ~2.3 million checks per day (~70 million per month/location) — we were already far beyond the free tier.
In July 2025, a sudden billing SKU change in GCP due to the vendor (DoIT) contract changes , the cost escalated for GCP uptime checks to ~$4,000 per day (~$120K/month, ~$1.45M/year), posing a direct threat to our operational sustainability and costs.
The Solution: Building Our Own Scalable Synthetic Monitoring framework using Instana Synthetics feature
Rather than absorbing these costs, we accelerated a proof-of-concept which was already underway to leverage Instana Synthetics as an internal alternative. Within weeks, we transformed this into a fully production-ready solution - migrating ~1600 Tenant Unit uptime checks off the GCP Synthetics and building a multi-region, automated monitoring framework entirely within Instana’s observability ecosystem.
During the transition phase in-order to keep the business continuity but to reduce the daily cost of uptime checks , we changed the frequency of ping tests from 1 min to 15 mins and reduced geo locations ( from 5 regions --> 3 regions) which brought the cost under 300 $ per/day from 4000 $ per/day as an interim workaround but continued to work on the long term solution.
Key elements of the solution implemented -
- Automation & Orchestration: Enhanced
instanactl and Jenkins pipelines to automatically create synthetic uptime checks whenever a new Tenant Unit (TU) is created. We used synctl -python based utility provided by Instana to create the tests.
- Code-Level Enhancements: Implemented changes in Janitor (Golang) so every new TU deployment automatically provisions uptime monitoring based in the region in which the TU gets created.
- Region-Aware Architecture: Deployed monitoring across 8 global environments (5 SaaS and 3 GitOps) aligned with TU placement.
- Real-Time Health Validation: Used lightweight ping-pong HTTP API Simple synthetic tests with OCM alerting for immediate detection and response.
- Interim Risk Mitigation: Reduced GCP check frequency and region coverage to stabilize costs during the migration phase.
This approach replaced a complex, vendor-driven system with a self-sustaining, deeply integrated monitoring framework under our full control.
Outcomes & Strategic Impact
What we Achieved and Why This Matters
-
We optimized our synthetic uptime check configuration to run from two strategically chosen Instana managed pop locations, instead of five in GCP and achieving the same reliable and consistent monitoring outcomes at a significantly lower operational cost.
Our Current cost in Instana Synthetics for running ~1565 tests across 2 managed pop location is : 27,048 $
v/s
Estimated cost is we continued using GCP uptime synthetics check for running ~1565 tests across 5 managed pop location is : 124,000 $
This initiative demonstrates how we can turn external challenges into opportunities for innovation. By solving a mission-critical problem internally, we strengthened Instana’s platform capabilities, improved cost efficiency, enhanced customer trust and built a foundation that will support future growth and innovation. It’s also a strong example of how our own product — Instana Synthetics — can deliver at enterprise scale, powering client-zero observability with resilience , precision and at lower price.
This solution is already delivering tangible business value and positions us strongly for the next phase of growth.
Screenshot 3: Example of one Tenant Unit HTTP API SIMPLE Synthetic test in Instana - Summary page
Screenshot 4: Example of one Tenant Unit HTTP API SIMPLE Synthetic test in Instana - Configuration page
#BusinessObservability
#Synthetics
#CaseStudy
#SRE