Case Study: Verizon achieves production-scale log management and massive Elasticsearch scalability with Elastic

Verizon - Customer Case Study

At Verizon, Senior Security Data Scientist Bhaskar V. Karambelkar led an effort to move logs off disks and into a searchable, production-grade log management platform using Elasticsearch. The team faced strict requirements — massive storage, very high ingest and query rates, schema agility, multi-tenancy and distributed fault tolerance — as volumes grew from ~10B total/500M daily in mid‑2013 to >500B total and 10+ billion daily events by December 2014.

They implemented a tuned, horizontal Elasticsearch architecture with dedicated master/data/client nodes, correct CPU–RAM–disk sizing, JBOD, bulk indexing, aliases, shard/replica and JVM/kernel tuning, query routing to client nodes, and integrated monitoring and retention (raw data archived in HDFS, ES retained ~90 days). The result was reliable, scalable ingestion and search at production scale — a multi‑tenant log management system running across dozens to 128+ nodes that sustained billions of daily events and enabled timely security analytics.

Verizon

Bhaskar V. Karambelkar

Senior Security Data Scientist