Tagged articles
write performance
10 articles · Page 1 of 1
Adding three secondary indexes to a MySQL 8.0 orders table caused write P99 latency to jump from 15ms to 2.3s and TPS to drop from 4800 to 1100 during peak hours; the article details a reproducible forensic process using A/B testing, invisible indexes, and sysbench to prove the culprit index had zero read benefit and safely remove it.
A/B testingInnoDBMySQL 8.0
0 likes · 23 min read

This guide walks through five systematic checks—indexing delay vs refresh, thread‑pool rejections, segment merges and I/O, bulk size and refresh settings, and mapping/_source configuration—to pinpoint and resolve Elasticsearch write‑performance bottlenecks.
ElasticsearchMappingRefresh Interval
0 likes · 7 min read

This article explains how indexes boost query speed but also incur storage, write‑performance, and maintenance costs, offering practical SQL scripts and guidelines for auditing, testing, and safely removing unnecessary indexes.
Database IndexesMySQLStorage Overhead
0 likes · 6 min read

This article explores various MySQL write performance tests using different concurrency levels—Statement, Connection, and database-level parallelism—provides benchmark results, sample Java/Groovy code, and discusses key configuration and architectural factors that influence write throughput.
ConcurrencyMySQLbenchmark
0 likes · 11 min read

This article evaluates MySQL single‑thread write performance by testing three insertion methods—continuous while‑loop inserts, JDBC batch processing, and multi‑row INSERT statements—detailing setup, code examples, and measured rows per minute to reveal their relative throughput and stability.
JDBCMulti-Row InsertMySQL
0 likes · 10 min read

This article examines the fundamental differences between B‑Tree and LSM‑Tree storage structures in distributed and relational databases, detailing their write and read paths, performance trade‑offs, update handling, lock conflicts, and high‑availability considerations to help engineers choose the right engine for their workloads.
B-TreeDatabase StorageLSM-tree
0 likes · 25 min read

This article presents a detailed performance test of HBase 2.x write throughput on a five‑node SSD cluster, identifies latency spikes caused by MemStore flush and ConcurrentSkipListMap size() overhead, and demonstrates how fixing the bug and applying in‑memory compaction dramatically reduces P999 and P9999 latency while preserving throughput.
HBaseIn-Memory CompactionMemStore
0 likes · 10 min read

This article experimentally demonstrates how increasing the number of secondary indexes on MySQL tables progressively degrades write operations—including INSERT, UPDATE, DELETE, and LOAD DATA—by measuring execution times across ten tables with one to ten indexes each.
DELETEINSERTLOAD DATA
0 likes · 14 min read

Didi’s engineering team analyzed a severe write bottleneck in their 3000‑node Elasticsearch cluster, identified long‑tail latency caused by refresh, translog locks, write queues and GC, and applied routing‑aware bulk writes, JVM and Lucene tweaks, and data cleaning to more than double write throughput while slashing server costs.
DidiElasticsearchLong Tail
0 likes · 17 min read

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
