Case Study: LivePerson cuts observability pipeline costs with Elastic on Google Cloud

LivePerson cuts observability costs by over 50% with Elastic

The customer, LivePerson, faced rising costs and performance constraints in its Google Cloud observability pipeline, which used Elastic's Logstash and Kafka. As the pipeline scaled, the use of general-purpose e2-standard instances for Logstash led to a compounding problem: more Logstash instances required a larger, more expensive Kafka cluster. The observability team at LivePerson needed to find a more cost-efficient infrastructure configuration but lacked a performance baseline to guide their decisions.

To solve this, LivePerson, in conjunction with vendor Elastic, implemented a structured benchmarking methodology to test five Google Cloud machine types for both Logstash and Kafka performance. The solution identified the n4d-standard-2 instance as optimal, which doubled Logstash throughput and cut the cost per event by over 50%, from $5.95 to $2.70 per 1,000 events per second. This efficiency gain from Elastic's approach allowed LivePerson to reduce its overall infrastructure footprint, requiring fewer Logstash instances and a smaller Kafka cluster, which significantly lowered costs and operational overhead.

LivePerson