PerformanceOptimization

Top Architect

Apr 18, 2025 · Backend Development

Optimizing XML-to-MySQL Import: Reducing Execution Time from 300 s to 4 s with Batch Processing and Multithreading

This article details how to dramatically speed up importing over 60,000 XML records into MySQL by profiling the original 300‑second implementation, enabling JDBC batch writes, configuring MySQL rewriteBatchedStatements, and applying a Disruptor‑based multithreaded pipeline, ultimately achieving sub‑5‑second runtimes with modest memory usage.

BatchProcessingDisruptorJava

0 likes · 15 min read

Optimizing XML-to-MySQL Import: Reducing Execution Time from 300 s to 4 s with Batch Processing and Multithreading

ByteDance Data Platform

Mar 6, 2024 · Databases

How ByteHouse Boosted Douyin’s Interest Circle Queries by 100×

This article explains how Douyin rebuilt its interest‑circle platform by replacing MySQL with the columnar OLAP engine ByteHouse, achieving roughly a hundred‑fold improvement in query speed, lower hardware costs, and seamless horizontal scalability for massive daily data volumes.

ByteHouseColumnarStorageDataWarehouse

0 likes · 10 min read

How ByteHouse Boosted Douyin’s Interest Circle Queries by 100×

Java Architect Essentials

Feb 23, 2024 · Backend Development

Optimizing Apache HttpClient for High-Concurrency Scenarios

This article explains how to improve a high‑traffic Java service by reusing a singleton HttpClient, enabling keep‑alive, configuring a pooling connection manager, and adding an idle‑connection monitor, which reduces average request latency from 250 ms to about 80 ms.

ConnectionPoolingHttpClientJava

0 likes · 11 min read

Optimizing Apache HttpClient for High-Concurrency Scenarios

Java Architect Essentials

Dec 25, 2023 · Backend Development

Optimizing SpringBoot Startup Time: Analyzing Bean Scanning and Initialization Bottlenecks

This article explains how to diagnose and dramatically reduce SpringBoot service startup latency by examining SpringApplicationRunListener phases, Bean scanning overhead, BeanPostProcessor timing, and cache auto‑configuration, then applying JavaConfig and starter‑based solutions to cut startup from minutes to seconds.

BeanScanningCacheJava

0 likes · 17 min read

Optimizing SpringBoot Startup Time: Analyzing Bean Scanning and Initialization Bottlenecks

DaTaobao Tech

Nov 24, 2023 · Artificial Intelligence

Performance Optimization of Depthwise Conv Int8 on ARM CPUs

By converting the input format to a C16 layout and exploiting the ARM V8.2 Sdot instruction, the Int8 depthwise‑convolution operator on ARM CPUs can be accelerated from 4.46 ms to 1.75 ms—a 2.5× speedup—though the required data‑rearrangement overhead prevents it from overtaking FP16 performance.

ARMDepthwiseConvolutionInt8

0 likes · 10 min read

Performance Optimization of Depthwise Conv Int8 on ARM CPUs

Bilibili Tech

Apr 28, 2023 · Cloud Native

Remote StateBackend for Flink: Design, Optimizations, and Cloud‑Native Migration

To enable Bilibili’s cloud‑native migration, the team built a RemoteStateBackend that moves Flink’s keyed state to the Taishan KV store, using deterministic KeyGroup placement, per‑shard snapshots, asynchronous write batching, off‑heap caching with Bloom‑filter filtering, and a fixed‑size memory model, which together reduce checkpoint overhead, improve disk utilization, and accelerate rescaling for more than one hundred production jobs.

CloudNativePerformanceOptimizationRemoteStorage

0 likes · 18 min read

Remote StateBackend for Flink: Design, Optimizations, and Cloud‑Native Migration

Code Ape Tech Column

Feb 20, 2023 · Backend Development

Optimizing SpringBoot Startup Time: Analyzing Bean Scanning and Injection Bottlenecks

This article explains how to diagnose and dramatically reduce the long startup time of a SpringBoot service by profiling the SpringApplicationRunListener and BeanPostProcessor phases, trimming excessive component scanning, manually configuring beans with JavaConfig, and understanding the impact of SpringBoot's auto‑configuration mechanisms.

AutoConfigurationBeanPostProcessorBeanScanning

0 likes · 18 min read

Optimizing SpringBoot Startup Time: Analyzing Bean Scanning and Injection Bottlenecks

Java Architect Essentials

Jan 31, 2023 · Big Data

Optimizing Large-Scale Data Retrieval: ClickHouse Pagination, Elasticsearch Scroll Scan, ES+HBase, and RediSearch + RedisJSON Solutions

This article examines a business requirement to filter and rank up to 100,000 records from a pool of tens of millions, presenting and evaluating four technical solutions—multithreaded ClickHouse pagination, Elasticsearch scroll‑scan deep paging, an ES‑HBase combined query, and a RediSearch + RedisJSON approach—along with performance data and code examples.

BigDataClickHouseElasticsearch

0 likes · 12 min read

Optimizing Large-Scale Data Retrieval: ClickHouse Pagination, Elasticsearch Scroll Scan, ES+HBase, and RediSearch + RedisJSON Solutions

政采云技术

Jul 12, 2022 · Big Data

Understanding Spark SQL Physical Execution Plans and Optimization Techniques

This article explains Spark SQL's physical execution plan, detailing each operator, how to interpret the plan, and practical optimization tips for data warehouse developers to improve SQL performance and resource utilization.

BigDataDataWarehouseExecutionPlan

0 likes · 10 min read

Understanding Spark SQL Physical Execution Plans and Optimization Techniques

DeWu Technology

Jul 8, 2022 · Big Data

Optimizing Large-Scale Product Set Refresh with RoaringBitmap

By representing pre‑and post‑refresh SPU sets as RoaringBitmaps and diffing them, the system avoids full‑insert writes, cuts memory usage by orders of magnitude, speeds refreshes by over 50 % and reduces write volume nearly 87 %, solving large‑scale tag‑based product refresh challenges.

BitmapDataStructureHBase

0 likes · 14 min read

Optimizing Large-Scale Product Set Refresh with RoaringBitmap

Baidu Geek Talk

Sep 22, 2021 · Mobile Development

Video Player Architecture Refactoring and Performance Optimization in an Android Short‑Video App

The article details how refactoring the Haokan Android short‑video client from a single global player to per‑holder multi‑player instances, combined with lifecycle‑lite event handling, pre‑loading, and dedicated thread operations, cut start‑up latency by ~150 ms, reduced frame‑drops dramatically, and boosted overall user experience and business metrics.

AndroidPerformanceOptimizationPreloading

0 likes · 20 min read

Video Player Architecture Refactoring and Performance Optimization in an Android Short‑Video App

DeWu Technology

Jun 4, 2021 · Mobile Development

Page Performance Optimization for Mobile App Webview

The DeWu app boosted WebView H5 page load speed by combining native‑level tactics—static JS/CSS bundling, HTML preloading—and web‑only strategies such as SSR, WebP image conversion, unused‑component removal, on‑demand lodash loading, and off‑screen image lazy‑loading, raising second‑open rates from ~5% to ~40%, improving Lighthouse scores by over 20% and cutting transferred bytes by roughly 20%, delivering an overall ~10% performance gain.

PerformanceOptimizationSPASSR

0 likes · 4 min read

Page Performance Optimization for Mobile App Webview

Didi Tech

Jan 12, 2021 · Big Data

Upgrading DiDi Real‑time Computing Engine from Flink 1.4 to Flink 1.10: Challenges, Optimizations, and Lessons Learned

DiDi upgraded its massive real‑time computing engine from Flink 1.4.2 to Flink 1.10, implementing a transparent migration across 1500 machines, adding native DDL, binary rows, MiniBatch, improved scheduling and window functions, and establishing a rigorous testing pipeline that achieved 99.9 % compatibility while preventing OOM issues.

BigDataPerformanceOptimizationRealTimeComputing

0 likes · 11 min read

Upgrading DiDi Real‑time Computing Engine from Flink 1.4 to Flink 1.10: Challenges, Optimizations, and Lessons Learned

Xianyu Technology

Oct 31, 2019 · Mobile Development

Performance Optimization of Alibaba Xianyu Fishpond Mini-Program

The Xianyu Fishpond mini‑program was optimized by keeping its instance alive, pre‑rendering first‑screen data, reducing costly setData calls, using $spliceData for long lists, staging tab‑switch updates, and integrating native video with cover‑view, which together cut loading time, eliminated scroll jank, and slashed tab‑switch latency, delivering markedly improved performance.

FrontendMiniProgramMobile

0 likes · 8 min read

Performance Optimization of Alibaba Xianyu Fishpond Mini-Program

Amap Tech

Aug 6, 2019 · Fundamentals

Off‑Screen Rendering Optimization for Car Navigation Maps Using Anti‑Aliasing FBO

By employing off‑screen rendering with an anti‑aliased Frame Buffer Object for stable map views such as the eagle‑eye mini‑map, the car navigation engine dramatically cuts CPU load and eliminates frame‑rate drops, reusing a single texture and following a concise FBO creation, rendering, resolve, and discard workflow.

CarNavigationFBOOffScreenRendering

0 likes · 8 min read

Off‑Screen Rendering Optimization for Car Navigation Maps Using Anti‑Aliasing FBO

Didi Tech

Jun 22, 2019 · Mobile Development

Mpx 2.0: An Enhanced Mini‑Program Framework for Cross‑Platform Development

Mpx 2.0 is an enhanced mini‑program framework that provides full cross‑platform support for WeChat, Alipay, Baidu, Toutiao and QQ, enabling developers to compile existing WeChat projects and native components to other platforms while offering sub‑package optimization, conditional compilation, setData efficiency and near‑zero migration effort.

CrossPlatformJavaScriptMiniProgram

0 likes · 16 min read

Mpx 2.0: An Enhanced Mini‑Program Framework for Cross‑Platform Development

Didi Tech

Dec 18, 2018 · Frontend Development

Mpx: An Enhanced Mini‑Program Framework for Better Development Experience and Performance

Mpx is an enhanced mini‑program framework that directly extends native mini‑program syntax with Vue‑style development tools, Webpack‑based compilation, reactive data handling, component‑level setData optimization, Vuex‑compatible state management, and team‑oriented features, delivering superior developer experience and performance for WeChat and Alipay apps.

DataBindingFrontendFrameworkMiniProgram

0 likes · 10 min read

Mpx: An Enhanced Mini‑Program Framework for Better Development Experience and Performance

Qunar Tech Salon

Dec 14, 2014 · Backend Development

Optimizing the SPR Library in LinkedIn Feed Mixer: Reducing Memory Usage and Latency

The article explains how LinkedIn's Feed Mixer uses the SPR library and presents several Java performance optimizations—including careful loop traversal, pre‑sizing collections, lazy evaluation, pre‑compiling regexes, caching, and cautious use of String.intern—to dramatically cut memory consumption and latency.

CachingJavaLinkedIn

0 likes · 8 min read