Speaker: Kishor Yadav Kommanaboina
Staff Software Engineer @Snapchat - Leading Caching and Storage Infrastructure
Kishor Yadav Kommanaboina leads caching and storage infrastructure at Snap Inc., where he architected the company's migration from KeyDB to Valkey across its caching fleet and built the autoscaling platform managing 800 Valkey and Memorystore clusters — roughly 20,000 instances across AWS and GCP serving 250M+ QPS. His work on write-heavy Valkey workloads, including full-sync failure resolution and dual-channel replication adoption, was presented at Unlocked Conference 2026 and published by the Valkey project. Before caching, he built Snap's telemetry ingestion platform — processing 1.8 trillion events daily — powering the company's analytics, experimentation, and ML systems. He previously worked at Amazon and Twitter.
Session
Autoscaling 800 Valkey Clusters: Distributed Systems Lessons from Scaling Stateful Fleets
Autoscaling stateless services is a solved problem. In a sharded datastore, a resize is not a capacity change — it's moving ownership of key ranges between nodes while the cluster serves live traffic, and rebalancing takes long enough that reacting to a traffic peak is already too late.