Md Shaghil, Rubriks
Tame Your Metrics Monster: Cut Observability Costs by 40% Without Sacrificing Reliability
Observability costs eating 30% of your infrastructure budget? You're not alone. Learn battle-tested strategies that deliver 40%+ cost savings while maintaining full visibility.
Discover how we tackled exploding metrics cardinality using three game-changing approaches:
"Dollar per Query": A simple framework that measures each metric's cost vs. actual usage, helping you identify waste and optimize ruthlessly.
Batch Metrics: Process non-critical metrics offline and load them on-demand in Grafana, cutting costs by 16% while preserving historical visibility.
On-Demand Metrics: Enable metrics via API only when needed for debugging or testing, saving 4% immediately with massive potential as we scale to production.
Plus quarterly audits that root out stale metrics and save another 16%.
You'll leave with practical guides: approach for usage-tracking dashboards, designing cost frameworks, creating self-service tools for engineering teams, and rolling out these solutions step-by-step. We'll share real successes, pitfalls to avoid, and how to empower teams to optimize their own metrics. Perfect for: SREs and platform engineers drowning in observability costs who need actionable strategies to reduce spend without compromising reliability. Stop letting metrics costs spiral—learn to make every metric justify its existence.
Md Shaghil is a Site Reliability Engineer with over 5 years of experience, currently working at Rubrik India Pvt Limited in Bangalore. He holds a Master's degree in Computer Science from the Indian Institute of Technology Bhubaneswar (IIT BBS). At Rubrik, Shaghil is part of the observability team, where he specializes in metrics services and cost optimization strategies. His work focuses on building scalable, cost-effective solutions that balance reliability with infrastructure efficiency.
