Sony LIV, one of India's largest OTT platforms, migrated its analytics infrastructure from a fragmented mix of batch pipelines, Elasticsearch, and BigQuery onto ClickHouse Cloud. The move consolidated telemetry from playback QoE, CDN performance, clickstream, and ad-tech systems into a single real-time platform, cutting query times from tens of seconds or minutes down to sub-second or low-single-digit seconds. The company reports improved incident response, better storage economics through columnar compression, and the ability to handle massive traffic spikes during live sporting events like cricket matches and Champions League games. Future plans include predictive autoscaling and deeper anomaly detection using ClickHouse as the foundation.

8m read timeFrom clickhouse.com
Post cover image
Table of contents
A stack built for batch, not live events #Finding a system that could do it all #A unified, real-time analytics architecture #Faster answers, better visibility, lower costs #What’s next for Sony LIV and ClickHouse #

Questions this post answers

How did Sony LIV improve analytics query performance for live streaming events?

Sony LIV migrated its analytics stack to ClickHouse Cloud, consolidating batch pipelines, Elasticsearch, and BigQuery workloads into a single platform. Queries that previously took tens of seconds to minutes now complete in sub-second to low-single-digit seconds, improving operational responsiveness during live sports events with billions of daily telemetry events. See how daily.dev surfaces real-world architecture stories for teams weighing analytics database migrations.

What database do large OTT streaming platforms use for real-time QoE and CDN analytics at billion-row scale?

Sony LIV uses ClickHouse Cloud to power real-time quality-of-service and quality-of-experience analytics across playback telemetry, CDN performance, clickstream data, and ad-tech events. The platform ingests billions of telemetry events per day through Amazon Kinesis and ClickPipes, replacing a fragmented stack of batch pipelines, Elasticsearch, and BigQuery. Developers evaluating analytics databases for telemetry at scale can follow similar case studies on daily.dev.

290 Impressions