How Trax Cut 75% of Kubernetes spend with PerfectScale

Trax faced strong efficiency goals from leadership but hit roadblocks optimizing Kubernetes clusters. Manual optimization using VPA and monitoring tools lacked clarity and was difficult to scale. The approval process relied on instinct rather than evidence, leading to engineering toil and service risks.

PerfectScale provided real-time, workload-level visibility across Trax's 200+ microservice environment. AI-guided intelligence enabled safe, evidence-backed optimization while protecting SLOs. The platform replaced an underperforming FinOps tool and provided granular cost details with actionable recommendations.

After eliminating wasted resources, Trax focused on identifying additional opportunities to improve efficiency and unit economics. The team delved into PerfectScale's data, seeking ways to meaningfully impact their cost-centric business metrics. A key metric for Trax is 'cost per processing,' which is heavily affected by their Kubernetes efficiency. PerfectScale has a unique feature that consolidates every replica of a service into a single view to provide a clear picture of the utilization trends across all replicas, which is especially useful for ephemeral workloads like Spark or Flink jobs. Trax leveraged this capability to understand better the heterogeneous utilization across the replicas of several of its heavily used services. This level of visibility helped Trax re-architect services to drive additional efficiency gains without impacting resilience or availability. They were able to build multiple flavors of the service with different levels of resources and route the incoming requests to the proper service based on the size of the data. This made a big impact on their 'cost per processing' metric. PerfectScale surfaced this data instantly, and without them, they would have spent countless hours evaluating hundreds of replicas to generate the same results.