✳flâneur — a map of the web's best reading

ClickHouse vs. Elasticsearch: The Mechanics of Count Aggregations

clickhouse.com · 3,447 words · saved by 1 readers

In another blog post, we examined the performance of ClickHouse vs. Elasticsearch on a workload commonly present in large-scale data analytics and observability use cases – count(*) aggregations over billions of table rows. We showed that ClickHouse vastly outperforms Elasticsearch for running aggregation queries over large data volumes. Specifically: Count(*) aggregation queries in ClickHouse utilize hardware highly efficiently, resulting in at least 5 times lower latencies for aggregating large data sets compared to Elasticsearch. This requires smaller and 4 times cheaper hardware for comparable Elasticsearch latencies. For these above-mentioned reasons, we increasingly see users migrating from Elasticsearch to ClickHouse, with customer stories highlighting: “Migrating from Elasticsearch to ClickHouse, reduced the cost of our Observability hardware by over 30%.” Didi Tech “This unleashed potential for new features, growth and easier scaling.” Contentsquare “ClickHouse helped us to sc

Introduction # In another blog post, we examined the performance of ClickHouse vs. Elasticsearch on a workload commonly present in large-scale data analytics and observability use cases – count(*) aggregations over billions of table rows. We showed that ClickHouse vastly outperforms Elasticsearch for running aggregation queries over large data volumes. Specifically: Count(*) aggregation queries in ClickHouse utilize hardware highly efficiently, resulting in at least 5 times lower latencies for aggregating large data sets compared to Elasticsearch. This requires smaller and 4 times cheaper hard

Explore this link on the map →

related reading