Elasticsearch cluster optimized from 12 nodes to 3, cutting infrastructure costs by 83%
Cloud & Infrastructure
Restaurant & Hospitality
Elasticsearch, AWS, Kibana
A search cluster that cost too much and buckled under load
Restaurant technology platform that processes high volumes of search queries across a large inventory of restaurant listings, reservations, and availability data. Their Elasticsearch cluster had grown to 12 nodes costing approximately $5,000 per month — but performance wasn’t scaling proportionally. Under heavy traffic, CPU and memory utilization spiked unpredictably, and the engineering team was spending time on firefighting rather than building product.
The root cause wasn’t the volume of data — it was how the cluster had been configured over time. Index mappings had accumulated without review, shard and replica counts were set to defaults rather than tuned for actual query patterns, and field types were misaligned with their use cases.
Index-level optimization and cluster reconfiguration not a replatform
Rather than migrating to a different search technology, we optimized the existing Elasticsearch deployment at the configuration level. The work focused on three areas: index mapping review, cluster topology tuning, and field-type alignment with actual query patterns.
We audited every index, identified field types that were incorrectly mapped (e.g., fields stored as text but only ever used for exact-match filtering), and restructured mappings to align with real-world usage. Shard and replica counts were recalculated based on data volume and query load. Refresh intervals were tuned to balance indexing throughput with search latency. Node distribution was reconfigured so the cluster could handle the same load with fewer, right-sized nodes.
83% cost reduction Zero performance trade-off.
The optimized cluster runs on 3 nodes at approximately $850 per month — down from 12 nodes at $5,000. CPU and memory utilization remained stable under the same heavy-traffic loads that previously caused unpredictable spikes. No queries degraded. No functionality was lost. The engineering team gained back the time they were spending on cluster firefighting.
$4,150/mo
~$50K/yr
75%
ElasticSearch integration, 96% API response improvement
In-house booking automation, 70% to 90%+ success rate
In-house booking automation, 70% to 90%+ success rate
React Native to native iOS, 100% session-preserved migration
