Douyin ElasticSearch Write Performance Optimization: Doubling Throughput and Reducing Write Rejections

By consolidating each bulk request onto a single shard, tuning node write paths, and pruning unnecessary fields, Douyin’s ElasticSearch team eliminated long‑tail latency, more than doubled write throughput to over 1 million ops per second, slashed write rejections, and saved millions in infrastructure costs.

Bulk RequestElasticsearchLong Tail

0 likes · 18 min read

Douyin ElasticSearch Write Performance Optimization: Doubling Throughput and Reducing Write Rejections

Xueersi Online School Tech Team

Sep 1, 2019 · Databases

Elasticsearch Data Write Process: Coordinating Node Phase Explained

This article explains how Elasticsearch handles data ingestion by detailing the roles of cluster nodes, the coordinating node workflow, index structures, shard allocation, request validation, pipeline processing, automatic index creation, and the underlying code paths involved in the write operation.

Bulk RequestClusterCoordinating Node

0 likes · 25 min read

Elasticsearch Data Write Process: Coordinating Node Phase Explained